/v/ - Video Games

AGDG: Angry gamedev general FUCK you mom I'm deleting the thread

>please don't shitpost this one to death

Good thing this isn't monster, because you've posted something with snouts

I will ask again before making my own.

There is no ready-made solution for texture atlas on Unity, right?

>>14569842

It seems there is not, at least, not one that lets you use height/metallic/smooth/normal maps too, rather than just the albedo texture

>>14569773

Sasha is a man's name, anon.

>>14569828

I deleted my own thread because they wouldn't stop shitting it up.

>>14569899

>>14569763

Speaking of (((something with snouts)))

>>14569828

Are you calling me a fucking goon? Maybe I should have made my greentext post of him "hurr I trol u" more fucking obvious who said what, because I forgot /v/ can't fucking read.

Christ. I'm out of here.

Christ anons let’s just get along and have a quality /agdg/ thread.

It really feels like we’re being consistently raided as of just recently though, and I take this from the perspective that this wasn’t really an occurance until we started having demo day (as otherwise it was mostly comprised of internal shit, and one off /pol/ lines that everyone just laughed at as it was contextually relevant to the discussion).

Hard not to prefer pre-DD /agdg/ threads, now they were comfy

Though it’s not worth taking as a degradatory sign to the quality of the persistent anons (u know who u are, I don’t always tho). As the randy randoms and such ignorables will get bored, move on, and we’ll still be here with a few quality additions who have a similarly thick skin. also our threads move way faster which it’s a bad sign

Okay, fuck it. I didn't want to ask, but I don't care anymore.

What have I missed? I'm not talking about the last thread. I've seen that one. What I'm talking about is that just about every thread either gets derailed or has one or two faggots trying. I've been gone for a while so I missed a couple of threads before and after the second demo day (including the demo day itself).

When I left, the /agdg/ threads were one of the few consistently high quality threads on /v/. The quality of these threads has taken such a nosedive in such a short period of time, that I wonder whether something happened while I was gone.

I'm noticing a pattern here. One or two retards bait with /pol/ shit, if you call them out, you get called either a Jew or a redditor, then the entire thread goes to hell and at the end the faggots claim that we should be grateful for having them around, because they keep reddit out, even though the threads were perfectly productive until they showed up.

Was there another incident similar to what happened back when ika released Red Sky while I was gone? Was there a major shitstorm and now we have an idiot with a messiah complex who thinks it's up to him, and him alone, to save us from reddit?

Take this for example: https://archive.fo/blPgz (post 14554996)

Look at the responses he gives the Anon who tells him he won't take the bait. Just look at how hard he's trying.

No, his theory is 100% correct. It's all me, they really should rangeban my entire ISP just to stop me from posting.

>>14570125

I’ve been pretty busy since around the first demo day so I’ve only had time to mobile lurk (almost every thread).

There’s been a more persistent pattern of (1)s and (2)s popping up in threads since that initial demo day; where our thread was sticked for 24hrs and the entire active population of /v/ couldn’t help but notice (in addition to the undesirables).

That first cyclical after demo day was consequently made a lot less “quality” compared to pre-DD threads by the prevelance of low-quality posts, but this pattern died off before the 2nd-DD. However, it re-emerged in the cyclical after the 2nd DD thread.

Though I’d like to mentioned that DD is just a catalyst for being able to see this happening, and shouldn’t be seen as “the” sole reason; as there’s other factors too.

Now DD is a good event, and it’s valuable to get critical criticism especially from the persistent /v/ users who are not /agdg/ inclined; as that’s valuable to improve our games, and is a meangful contribution from /v/.

Though, it has introduced as consequence, an element that arises from too much visibility in this type of discussionary format. From this is the establishing of a meta-aspect of culture forming around the persistent set of people actively contributing; generally in a manner that falls under the “official” and “unspoken” rules.

So, we can see that there’s a new active element that doesn’t understand said rules of the culture, and even maliciously breaks those rules intentionally or are subconsciously driven to do so (what value this has in their life, and how it’s meaningful to them is questionable; though it is unquestionable that they have serious personal issues they need to resolve).

>>14570157

I understand why u say that, but it’s an identifiable pattern which can be articulated into a more concrete form.

So it’s more of an interpretation of observations; so a hypothesis.

>>14570370

>maliciously breaks those rules intentionally or are subconsciously driven to do so (what value this has in their life, and how it’s meaningful to them is questionable; though it is unquestionable that they have serious personal issues they need to resolve).

>breaking implicit rules on a general forum thread means someone has serious personal issues

Imagine being this autistic.

Is the Source engine a good one to start with?

t. aspiring ideas guy

>>14569770

Is that Playmaker, Bolt, or….?

>>14570456

Source is a good engine for slow pcs.

>>14570399

Again you’ve slipped up, and are showing your true colors; as you know that statement is subconsciously true.

My honest advice is to grow some self-awareness by attending to your self, as an outside observer to your personality this is really embarrassing to see someone so lacking is self-awareness, and do something meaningful instead of spending your time contributing nothing of value.

Hell, start debating for real in a context you care about, and get into some intellectual arguments; just improve yourself man, as this is really sad, and I pity you.

Notice: you are now filtered, and all of your following IDs will be filtered; as you have nothing of value to say.

>>14570483

>Notice: you are now filtered, and all of your following IDs will be filtered; as you have nothing of value to say.

Announcing filters is for drama queens.

>>14570483

>Again you’ve slipped up

As an outside observer, let me tell you that psychology is a pseudo-science and trying to perform psychoanalysis on random anonymous people online is even sillier.

>>14570497

Psychoanalysis is surely psychiatry not psychology.

>>14570500

It's a pointless distinction, but even if you use the kike's definition it still falls under "psychology".

>>14570491

I thought it was relevant to the discussion of thread quality (discussion I was attending to), and that it’s a relevant action to set a precedent for the undeseriables; though yes I agree that it’s generally rightfully considered unnecessary drama.

>>14570503

Select the corner edges (the perpendicular edges you want to tighten) and hit ctrl-e and drag that shit up to 1.

or if you are not using blender make extra edge loops around the corners

Anons, whats your favorite 3d software for modeling characters?

>>14570546

You’re interpreting that line a bit too axiomatically, and defining posts such as yours as undesirable; which they’re not, it’s valuable discussion of an albeit meta topic.

So it’s more like “if you break rule 8, and don’t contribute anything of value; then you are lacking value as an anon”.

So, my question is then, why should I waste my time with someone matching that; who provably lacks value in the context of an /agdg/ thread?

Also it’s relevant to mention that tyrannical censorship by powerful entities is unethical, but I’m certainly not a government doing so not am I someone who possesses administrative power here; i’m just a single anon using the tools at my disposal.

>>14569993

Quality article. Time to get back to reading it.

I hope that stalker anon decides to takes inspiration from that as it’s has really interesting gameplay implications.

>>14570606

Blender is pretty great, but hotkeys are annoying to memorize; though once u get it down the hotkeys are pretty convenient/fast compared to being menu centric like maya.

I’d reccomend to use Zbrush or something similar for detail work like normal maps and the like (if u want that level of detail, that is).

>>14570667

another vote here for Zbrush. 3D Coat is pretty cool too, so I hear.

Blender sculpting function is buggy as shit. Avoid.

>>14570695

And so am I, but that doesn't mean it isn't the cause of the problem here.

It's just an vaguely relevant image to illustrate my disgust.

>>14570667

I haven't played Stalker yet so I was surprised to see its AI shared some similarities with AIDS Berserk. Having a separate Windows partition instead of just using Wine means I play Windows exclusive vidya less than I used to, since I do almost everything in Linux these days.

>>14570689

Stallman has no problems with software making a profit as long as it remains free as in freedom.

>>14570380

You could fake it with props, but a fully dynamic destruction engine would need a monumental amount of work. Some mappers have attempted to fake it with func_physboxes with limited success. It doesn't perform well for low end cpus and it's not feasible to run in multiplayer.

>>14570714

What would that be?

>>14570755

speebot, although I think it got maybe 5-10 sales total

>>14570667

I agree with that about the hotkeys, i wanted to change to another 3d software but most of my progress it's in there, maybe next time i'll give a try to 3Dcoat.

>>14570678

The sculpting mode of blender it's weird af, i give a vote to Zbrush too, maybe i'ts better use softwares that are aimed to a certain funtion, they make things considerably fast and easy.

>>14570689

Linuxfags please leave.

>>14570755

Speebot

>>14570803

>Implying because he's a linux fag

Linux fags make tons of money via donations, speeches, etc. That dude is just a fucking idiot.

>>14569770

This seems to have come along a bit since I last was here.

>>14570887

Anyone bitching about /pol/ posts is probably cancer from reddit or cuckchan who doesn't even know why most of us are here.

>>14569773

wheres the bellybottun?

>>14570960

suppose you could try beveling the edges

>>14570835

The GPL is unreasonable for commercial software as it allows anyone who buys it to give it away themselves. I think you're holding the code to a different standard than the art/music/levels of a game, when really they are all works that can be sold the same way. Why should the code be free, and not the art? It's like saying that an artist can sell their work, but a programmer must give everything they make away for free.

>>14570979

I tried that too and it didn't work.

I'll just take a break for now and look up tutorials later. That or I just remodel some of these parts. It would probably work if I make that thing a hollow tube.

And sorry for calling you an asshole

Im thinking about using spine2d to make my game sprites, and in fact already bought the license. The sprites created by it seem to be easily manipulated if I need to change something down the road. Do you think the sprites will look good in a metroidvania style game?

>>14571013

Tweening is THE defining feature of flash garbage, and turns even good artists into korean-tier mass producers of shit. You're also an idiot if you actually bought a software license. This is just my opinion, of course, but you should know that everyone else also agrees with me.

>>14571003

STALKER and Half-Life 2 do exactly what he mentioned

>>14571011

its fine, i was mad at my inability to grab the edges that i wanted when i named it

>>14571032

The source engine is not GPL'ed in any way , shape, or form. What are you talking about?

>>14571060

>do exactly what he mentioned

That dude never mentioned GPL. He simply said the engine is open source and x-ray and source both have their source code on github.

>>14571145

Are there any successful games with a large budget that went FOSS within a year of release?

>>14571060

>The source engine is not GPL'ed in any way , shape, or form. What are you talking about?

Huh??? I thought it was called the source engine because it was open source.

>>14571090

Having your code on github, or otherwise publishing the code, does not make it open-source. This is what open source software is:

https://opensource.org/osdhttps://opensource.org/osd

The source SDK is on github, the source engine is not on github and is not publicly available. If we look at the SOURCE 1 SDK LICENSE right on valve's github, we can see that from the first sentence that it is not open source:

>You may, free of charge, download and use the SDK to develop a modified Valve game

running on the Source engine.

Which directly violates point 8 in the open source definition.

Link: https://github.com/ValveSoftware/source-sdk-2013

As for X-Ray, this is obviously not open source either, see:

>any changes to this engine are allowed for non-commercial use only

Which directly violates point 1 in the open source definition.

Link: https://github.com/revolucas/CoC-Xray

Open source sounds good in theory- but when you read the actual definitions and licenses, it's asking you to give up a lot more than you realize. That's why I don't think that it's reasonable, and it's also why both of the examples that you mentioned are not licensed under open source licenses. They don't think it's reasonable either.

>>14571145

Software does not have to be free for others to contribute fixes and improvements to your code. Putting your code under a free licenses such as the GPL will do this, but it's not exclusive to the GPL by any means. And it is exactly like demanding that programmers give everything away for free. Just because they might get donations if people feel like it doesn't mean that it's not being given away for free. You could make the same exact argument for art. Why shouldn't artists just give away all of their art for commercial use, and then beg people for donations instead of charging money?

>>14571196

What you described is FOSS.

>>14571203

This is what happens when a bunch of retards repeat something over and over; Sooner or later a word gets "repurposed" and then regular people start following that new definition because they want to be correct. A few years of that and the word actually does mean that. guess which stage we're at

>>14571233

Whats the difference between FOSS and OSS

>>14571248

OSS = You can see/access the source-code

FOSS = You can see/access the source-code and are free to do what you want with it

All FOSS is OSS, not all OSS is FOSS.

>>14571248

Open source: A term that formed naturally to describe software whose original source code is publicly available

FOSS: A term specific crafted by autists to describe a specific subset of open source software

>>14571203

No, I am not describing FOSS. Open source software is not free software. The page I linked too is from the Open Source Initiative which is NOT the Free Software Foundation and does NOT describe free software, even though they might have a lot in common.

Please, see the 'Common Misunderstandings of “Free Software” and “Open Source”' section of this page for more info:

https://www.gnu.org/philosophy/open-source-misses-the-point.html

>>14571233

This is why we are having the problem: the anon(s) who are posting the Stallman pictures are demanding software to be free and open source on the kinds of terms that are published on sites for groups like the Open Source Initiative and Free Software Foundation. Meanwhile, to someone who isn't familiar with how FOSS/OSS advocates use these terms, "Open source" sounds innocent enough, because they don't know what we are being asked to give up by making our code "open source". So when I argue against FOSS/OSS advocates, I sound unreasonable to people who aren't familiar with the fine print of the "official definitions" they put out.

>>14571258

>>14571259

When someone who isn't a FOSS/OSS advocate uses the term, that's what it means. But when you see someone posting Stallman pictures, the open source initiative link is what they are talking about. They do NOT just mean making the code publicly available. They want you to license it in a way where anyone can redistribute your work and sell it for money, without asking you, as well as all of the other things on the list here: https://opensource.org/osd

>>14571282

>no, hellbreaker doesn't fucking count, seriously did anyone ever see him post about his game even once?

No, but I've been repeatedly assured that he posts here all the time

wait a second

>>14571290

He only very rarely posts, and probably just participates to the level where it's acceptable for him to post his trailers and steam page here.

>>14571282

>hellbreaker doesn't fucking count, seriously did anyone ever see him post about his game even once?

I did. I was one of the play-testers and was invited to playtest via these threads.

>>14571185

If you check https://en.wikipedia.org/wiki/List_of_commercial_video_games_with_available_source_code , most commercial games which went FOSS did so after a few years, while most ones that did go FOSS within a year were lower budget indie titles. If you count games where the source code was released under a year through proprietary modding toolkits or under proprietary licenses, that number rises much higher.

>>14571196

And is there something wrong with giving away software for free? You can still make a profit from it and many developers do. Comparing code to pictures is a bit silly too since they're used and modified in different ways.

>>14571282

>>14571290

Hellbreaker anon posted more on the /agdg/ board than the /v/ threads, and often had long periods of silence. Missing him is understandable since many anons here probably ignore the board.

>>14571319

I say a year because I get the feeling most games with a budget that release the source code (which isn't FOSS or OSS apparently) do so after they've made all their sales. The first one that seemed big that I noticed was aliens vs predator. When I looked it up I saw in wikipedia it mentioned "In 2001, on end-of-support and end-of-life of the game". Almost all of those games listed release their source code years after the game is dead.

>>14571325

There isn't anything wrong with making the choice to give your stuff away for free. But I really hate to see the kind of double standard that people have with software and art. Nobody goes around asking artists to release their art assets for free commercial use. But then when it comes to code, you have posters like >>14570835 who believe that code should be released for free, and go around asking people to just give the commercial rights to their code to everyone, like it's expected of them. It's a terrible double standard and I don't think it's silly to compare code and art in this regard, they both take a huge amount of work and skill to be able produce good programs and good art assets, but one of them is treated as if it's something that nobody should have to pay for (as if it has no value in and of itself) while the other does not have this expectation that you should give away your work.

>>14571351

/vg/ has a cyclical too

>>14570689

That's bullshit. The vast majority of people here since I started lurking years ago have always been in it for profit or for a career in the long run. And there is nothing wrong with that.

>>14570953

Now that you mention… i forgot

Gonna fix it tomorrow

>>14569896

>google "sasha"

>go into images

>its only women

Not really

>>14571533

>google "white inventor"

>go to images

>its only black men

hmm

>>14571537

Well fuck

I don't know why Unity vectors aren't just represented as an array. You get retarded shit like this.

	public float this[int index]
	{
		get
		{
			switch (index)
			{
			case 0:
				return this.x;
			case 1:
				return this.y;
			case 2:
				return this.z;
			case 3:
				return this.w;
			default:
				throw new IndexOutOfRangeException("Invalid Vector4 index!");
			}
		}
		set
		{
			switch (index)
			{
			case 0:
				this.x = value;
				break;
			case 1:
				this.y = value;
				break;
			case 2:
				this.z = value;
				break;
			case 3:
				this.w = value;
				break;
			default:
				throw new IndexOutOfRangeException("Invalid Vector4 index!");
			}
		}
	}

>>14571471

>comparing source code to art assets again

Software is not like most artfag content. As new computing platforms (both hardware and software) arise and bugs or security flaws are uncovered, it runs the risk of becoming outdated, dangerous, or even flat out unusable unless you hunt down aging hardware and software to run it or find an emulator. The internet is littered with dead proprietary projects with no chance of revival and most of the still active programs are effectively black boxes. Having problems or wanting to add new shit? Guess what, unless you write your own implementation from scratch you're probably screwed.

FOSS and OSS projects still die, but any third parties can revive or fork software at will. There's much less duplicated effort than the splintered proprietary space and usually less retarded spaghetti code since your shit is open for the world to see.

Video games are one of the final software industries where proprietary software has a stranglehold. This is partially because programmers avoid its shit pay and working conditions, leaving the industry full of shitters. Even when the developers are competent and contribute to free software outside work, publishers regularly push them around and force them to add unwanted DRM or prevent them from releasing ports for example, there's several EA and Zenimax studios which internally port their games to Linux but the publishers won't let them support anything but Windows, consoles, and maybe macOS.

I'm confident video games will move towards free software as well, this shit just takes time and the vidya industry is notoriously slow to adapt.

>>14571656

And here I thought their matrices couldn't be worse

	public float this[int index]
	{
		get
		{
			switch (index)
			{
			case 0:
				return this.m00;
			case 1:
				return this.m10;
			case 2:
				return this.m20;
			case 3:
				return this.m30;
			case 4:
				return this.m01;
			case 5:
				return this.m11;
			case 6:
				return this.m21;
			case 7:
				return this.m31;
			case 8:
				return this.m02;
			case 9:
				return this.m12;
			case 10:
				return this.m22;
			case 11:
				return this.m32;
			case 12:
				return this.m03;
			case 13:
				return this.m13;
			case 14:
				return this.m23;
			case 15:
				return this.m33;
			default:
				throw new IndexOutOfRangeException("Invalid matrix index!");
			}
		}
		set
		{
			switch (index)
			{
			case 0:
				this.m00 = value;
				break;
			case 1:
				this.m10 = value;
				break;
			case 2:
				this.m20 = value;
				break;
			case 3:
				this.m30 = value;
				break;
			case 4:
				this.m01 = value;
				break;
			case 5:
				this.m11 = value;
				break;
			case 6:
				this.m21 = value;
				break;
			case 7:
				this.m31 = value;
				break;
			case 8:
				this.m02 = value;
				break;
			case 9:
				this.m12 = value;
				break;
			case 10:
				this.m22 = value;
				break;
			case 11:
				this.m32 = value;
				break;
			case 12:
				this.m03 = value;
				break;
			case 13:
				this.m13 = value;
				break;
			case 14:
				this.m23 = value;
				break;
			case 15:
				this.m33 = value;
				break;
			default:
				throw new IndexOutOfRangeException("Invalid matrix index!");
			}
		}

>>14571656

Basically it boils down to how C# handles things.

Arrays are objects with reference-type semantics. I can't remember if they're allocated on the heap or the stack, but they're more expensive than a primitive type. Likewise, since it's a reference, it's not stored in contiguous memory (an array of arrays could be all over the place).

Meanwhile, structs are more dense for data and their allocation is more "immediate" than a reference type, especially when created in limited scopes (eg a local variable in a function call). Also, because they aren't referenced, the object graphs are simpler, making it much easier for the GC to do its job.

Unity's approach here is actually okay for what it does - it allows for array-like indexing but still implements a vector as a struct (data) type.

>>14571675

XNA/FNA and a few other frameworks in C# that I've worked with also do the mXY pattern for matrices. It's ugly, but it's necessary for performance. There's no more atomic way to respresent it.

>>14571688

>>14571692

Guess I'm wrong.

>Also, because they aren't referenced, the object graphs are simpler, making it much easier for the GC to do its job.

is this a good time to bring up that unity's implementation of a GC only has a single generation?

>>14571699

Why the fuck does Unity have a specialized GC? Oh well.

Also I guess it's not fair to say that struvts are "faster" for GC. Rather value types with only primitives (eg vector floats) won't contain references so they're endpoints of the GC object graph which makes them simpler in that regard.

>>14571714

>specialized GC

That's one way to put it, I think they've just been using a several decade old open source GC because they don't want to license or implement a newer one. Blog posts must take too much of their time. bdwgc supports generational GC so it still doesn't explain why they don't use it.

A generational GC should be obvious, because it's clear to everyone that Unity generates a lot of trash :^)

>>14571657

Being able to edit and modify the source is fine. That's not what this is about. I'm fine with people being able to read, modify, and distribute my code. I just don't want people to make money using my code (since it's my work) without paying me some of the money. Unfortunately the FSF/OSF don't believe in that- and this incompatibility is written into their licenses- so I can't agree with the expectation to release my software as free software or open source software. I will release my code in a way similar to this:

https://raw.githubusercontent.com/avoitishin/xray-16/master/License.txt

Which is not free and not open source, yet it does not have the issues that other proprietary software has. Anyone can fix bugs and redistribute those bug fixes, and anyone can port that engine to other platforms.

>>14571282

> I'm saying that they are faggots for one reason or another, and coincidentally the one thing they all have in common is that they plan to sell their games.

Is it not possible that you simply begin to recognize people the more they post? There is bound to be some faggotry if you look closely so the end result is that you start to see faggots everywhere.

Just a thought.

>>14571810

On second thought I probably should really have used insets…

I decided to test the speeds of summing 100,000 fake Vector4's, one based on an array and one based on XYZW and unity fucking crashed

>>14571870

What do you mean anon?

Adding random values in a Vector4[]? Or taking a single Vector4 and using the indexer that many times? It'll be slower than directly setting the value yourself

>>14571870

>>14571890

https://hastebin.com/wegevirewe.cs

So I made a very naive implementation that just holds the values, plus used an (older) benchmark class that I made. Basically you give it a delegate and it performs it C number of times, performing N iterations per test.

In ticks, with 20 iterations at 1m values per iteration:

>r1 = 13841623 (setting fields)

>r2 = 21074888 (setting properties/indexer)

>r3 = 10374596 (constructor/new object)

Therefore, it's preferable to set the values yourself, directly, Properties are much slower to set data, since it involves the overhead of a method call. Thus, value types commonly expose freely modifyable fields directly, for performance (as well as passing by ref/out to method calls, or passing an array).

This is a design no-no according to C# specs: structs should be readonly, only set via a constructor, and class values should only be publicly exposed through a public setter or method call. Of course in practice, value types are freely modifyable for performance, and use of properties depends on the programmer.

>>14570758

>>14570815

>Speebot

I though it would be Risk of Rain?

>>14570441

it's custom made

>>14572002

That came out of 4chan's /agdg/, calling it "this place" is stretching it a bit.

what happened to the old thread?

>>14572485

the op got butthurt over goons not liking his easter animu image so he deleted it

>>14572583

Analytics as much as I hate them are actually really interesting. I posted a link to my thing in 8chan, and sometime later when I looked at the analytics there were visitors from 4chan and some weird forums.

Does anyone know if ue4 packages personal data into a project, in a way that someone other than epic can extract it?

>>14572627

Define "personal data".

>>14572583

None of those but YouTube surprise me. Why yt of all places

>>14572634

Probably a YT video on some political subject with a link to a /pol/ thread in the description,

>>14572633

well something someone could use to identify me or the pc used for building it.

>>14572634

if the owner of the channel doesnt care youtube comments are probably less moderated than most imageboards, so some comment sections are pretty much /pol/

>>14572623

a demo day thing, perhaps? that IS interesting

At least you got publicity

>>14572640

Okay, that's a very broad definition of "personal data". I have no way of knowing whether or not they have something like that built in. However, given that Epic takes royalties and they need a way to enforce it, I would assume so. They probably have something in there to help them prove it in case of a legal battle.

>>14572641

It was a stupid app that mirrors images. Either way it's very interesting to see what kind of places it spreads into.

>>14572634

I usually have AvE, the occasional tutorials if I'm learning something new or just a podcast running on the second monitor. To me, at least, Youtube is the least surprising one.

>>14572646

Thats why i said someone other than epic. Im sure they have something in there i just want to know if someone just trying to fuck with me is going to get to it. Im also not banking on them not giving out my info if law enforcement asks them.

>>14572657

Sorry, didn't read properly.

The same problem as with the "security through obscurity" principle applies: if the mechanisms are there, they are there and can (at least theoretically) be used by anyone.

Whether that's a problem depends on what sort of mechanism we're talking about and I already told you that I sadly cannot tell you. I wish I could help you, because it's an interesting question and I use UE myself, but I can't.

If you're that paranoid about this, I suppose you could set up a separate Windows installation for developing your game. You know, without any of your personal data.

>>14572303

True enough.

Anyone else use a proper planning tool? I've started using hacknplan for a bit and I really feel it helps me keep track of my progress and not get overwhelmed by shit that needs to be done.

>>14572775

>developers trying to be funny

I wish they'd stop.

>>14572800

Yeah, it's almost like there's a reason for why they're developers and not comedians.

>>14572775

Looks pretty nice, do you work strictly by kanban or just using the board?

>>14570887

It's like you don't even know where you are. Stop debating "pol posts", start agreeing with them and moving on. That, or GTFO redditfag.

>>14572819

I added and categorized everything to the kanban and epics. I always pick a couple of task I wish to focus on over the next days/weeks. When a task is assigned to you, it appears in a todo list on your profile's dashboard outside of the project page itself. That's a really nice feature, because all of the other things that need to be done aren't listed there. So you get a concise todo list that isn't cluttered by unrelated tasks.

>>14572854

this is /leftypol/ trying way too hard to make /pol/ look bad

>>14571656

The further from C (or assembly) you get, the more cancerous coding becomes. The idea that a language wouldn't have an indexed data primitive stored in contiguous memory when allocated on the heap is a road sign for you. It's a sign that says you've gone too far. You must go back.

>>14572880

Damn. My, "Listen and believe" reference was subtle enough?

>>14572920

>/v/ always made the best to stay away from /pol/ bullshit, and that's the biggest indicator you might be from somewhere else

Holy shit, the resident redditer is telling people they're outsiders if they don't "stay away from /pol/ bullshit". Let me guess, that includes thoughtcrimes like racism?

>>14572943

>i think you're exaggerating a bit

We do have several active reddit crossposters. And at least one of them is sperging out about "casual racism" at any given time.

>>14572962

nigger

>>14572965

if i call mark a fat kike is that casual as well?

>>14572943

>i think you're exaggerating a bit

this is rotten redditer and he's literally posting his reddit caps in this thread, I think it's safe to say I'm right

>>14572958

i generally like the idea of shipping a game with the source code in it (or optional)

at least then i won't have to bother making mod tools

>>14572973

>kike

Wow antisemites are not welcome here okay?

>>14572973

yes

>>14572962

take for example our resident /share/ thread with animu hittler pics

that's different from being actually retarded like >>14572975

just look at his post history

>>14572976

>source code included with purchase

I like that idea. I've planned (if I ever release a game) to sell my game, but it'll be open source and free to anyone who can compile it themselves. I think asesprite does something like that.

I'm not a fan of niggers myself but I don't come to /v/ to listen to some faggots whine about them. This is not the place to discuss niggers or the decline of western civilization of whatever. And for those whose argument is that closing your eyes doesn't remove the problem. You're absolutely correct. But not wanting to talk about it here, with you, does not mean I'm closing my eyes. Video games and casual shitposting, not autistic screeching about niggers and kikes, praise vidya.

>>14573159

< What is the difference between transform.basis and transform.origin?

>he hasn't spent years failing geometry and algebra exams only to learn minor shit about vectors

i don't know if this is what they're supposed to do in godot, but as i see it, origin would be where (0,0,0) is for that transform in the global coordinate system, while basis would be the directions

example being a regular three dimensional space

origin vector is (0,0,0) - treat it as the local position of a transform instead of global

directions are (0,0,1),(0,1,0),(1,0,0)

>>14572775

This seems better than what I’m using atm, I’ll switch to this method and see if it works out, thanks for the info

I keep hearing about a thing called a design document, is this a useful tool for keeping track of what you need to do?

>>14573385

As long as you don't let it bloat.

>>14573424

>you're never gonna finish it anyways

y-y-you don't know that, baka!

>>14573385

You're going to find the vast majority of things for managing and documenting software development (like UML) are shit. Be extremely sceptical about shit like that, because the majority of the time those things start out making sense, until corporations get wind of it and they just go to complete shit. Just do what makes sense, don't mindlessly follow industry buzzwords.

>>14573447

>i started aggydaggy breads on this tapestry years ago and i have yet to finish a single game

Jesus. Go back and shit out a tetris clone and work from there.

>>14573495

>that NPC dialogue

Welcome to the coalition anon, might as well start using coke.

>>14573521

This game, the laughing is intensifying to a point we may not be able to contain it.

>>14572962

When you call someone a gay nigger in a casual conversation.

>>14572997

Yes, Aseprite is free to anyone who can compile the code themselves. A friend of mine who thought he was computer literate tried it and ended up paying the $15. It's why the plebbit post in >>14572958 is absolutely right. Just because the source code is out there doesn't mean that you'll lose money on the game. In reality it probably just means better mods because the end user has full-access to the game's code. Which I'd like to do for at least some of my future games because I love seeing things like people heavily modifying Doom to make a First-Person RPG.

>>14573424

>>14573385

It's a nice alternative to keeping hundreds of scattered notepad documents everywhere. I just use a simple Word document, with it's automatic header numbering system, so I could quickly know what page something is on if I need something specific. I'm in very early stages, too, so it helps to write down stage ideas, then expand upon them if I can. I also include puzzle designs, specific missions and so on. It's 18 pages so far, but only 12 of so are full of actual notes, the others are puzzle designs and so on.

>>14573385

I use Zim desktop wiki when I need to organize a large amount of text and Trello for keeping track of tasks.

>>14573597

What about what happened with Lugaru, where someone used the source code to release their own HD version, earning money off someone else's game?

>>14573636

considering that overgrowth is still in development, i think he kind of deserves it

>>14573660

but overgrowth was released a while back

>>14573663

my point is the same, 9 years development and you expect people not to do a better job than you?

>>14573660

Doesn't really help someone when their game is stolen.

>>14573713

>re-releasing someone else's game is not stealing

>>14573739

>if it's a crime, he can always sue

>if he hasn't sued yet, then it isn't really an issue worth discussing

Because suing is so cheap. So cheap, in fact, that a common tactic of AAA is to bankrupt competition with a false lawsuit.

>>14573750

yea, but we're not talking about AAA here

if i steal your bike and we're both poor shits, it doesn't mean that i get to keep your bike because neither of us can afford lawyers

>>14573750

Yeah, if only there were some sort of government agency going through every file on the internet to determine whether or not something's stolen.

>>14573762

>He can sue even though he can't afford to sue

>>14573771

>Many plaintif's attorneys will work on a commission basis: their fee will be a percentage of what is collected from the person being sued.

and if it's not enough to pay an attorney, then it's really too low of an issue to even care about

>>14573636

That's basically the same as downloading the binaries to someone else's program, cracking it if it usesDRM, and reselling it on another store. You don't need the source code for that, scammers were doing it on Itch.io for a while and given how poorly curated Apple's App Store is I'm sure it's been done multiple times before on there too.

tl;dr that's a store curation problem, not an open source with proprietary assets problem.

>>14574041

Texture arrays could always work too, and you could use a similar lookup routine.

Seems u have pretty much all the bases covered, but u could stand to make the lookup texture simpler to use; like packing the 2d lookup corrdinates into a single int.

Also, you could just have 3 atlas’, and just use the same coordinates for the albedo/normal/etc.

You may want to look at tri-planar shading as with using world coors for UVs you get stretching on different axis.

>>14574094

Also if u pack the lookup corrdinates in a single int all you need is a 1d tex for lookups in the shader, and it’s be easy to use.

>>14574042

Looking good.

>>14574042

The one thing that really pissed me off in Speebot was that the buildings cut off right at the water, that really makes me mad.

>>14574094

>>14574099

good to know it actually makes sense and just need some adjustments

I also meant tri-planar but didn't knew the name, so went with whatever I know it's associated with it

I actually thought that it was the correct name when I used world coordinates

>>14574131

Not sure what you mean? The buildings go way below the surface, it's just that the water is not translucent because of the art style.

>>14574138

i've only played the demo, but most of the time i had trouble judging the distance when jumping vertically

>>14572854

Did you read my post? I'm literally telling the guy to stop debating /pol/ posts and fuck off.

>>14572920

>>14572958

You're literally from reddit and you're now trying to call people who have been here for many months and years outsiders. Get the fuck out and stop ruining our threads. I know that you haven't started posting on here again until about two months month ago, because ever since you've started doing that, these threads have been consistently run into the ground. I can't understand why the moderators haven't permanently banned you at this point.

>>14573448

This is not any different from saying that nobody will buy your game because they can download any number of free games that have had millions of dollars put into them. Yet, people would say that it's unreasonable to ask everyone in this thread to release their games for free. But now that it's software, people can argue for it without realizing the audacity of what they are saying. I doubt anyone else but me will use m y game engine any way. I want to make my own games on the engine, and that is about it.

>>14574235

>Yet, people would say that it's unreasonable to ask everyone in this thread to release their games for free.

But anon, every game is free.

>>14574235

>you're now trying to call people who have been here for many months and years outsiders

he really doesn't seem to be doing that though

>>14574245

Yes- but putting it on a store makes more people feel like buying it, then asking for donations.

>>14574231

The only approved placement of const is in the garbage can.

>>14574288

you dare defy scott?

>>14574041

What you want to do is probably better implemented either by:

1. Fetching data from a uniform buffer. (If you want a unique access per vertex or fragment shader invocation)

2. Supplying extra vertex data. (If you want a unique access per vertex.)

3. Supply vertex data through instancing. (If you want a unique texture per vertex instance invocation)

Textures are meant primarily for data you want to "sample", applying filtering techniques. The data you want is just raw data, so you should avoid using textures if possible.

That said, whether that's possible all depends on your engine's support. If you're building your own shit or the above are supported, then consider it.

The main question you should ask is how you actually want to use it. Please go into detail of why you're going down this madness of indirectness.

>>14574094

>Texture arrays could always work too, and you could use a similar lookup routine.

I want to add here that these MUST be texture array objects, and not arrays of textures. Which texture you access must be dynamically uniform, so trying to access tex[i] is illegal and will probably produce artifacting, assuming your shader compiler even accepts it at all.

>>14574235

Unless you make some disposable mobileshit game, games have some qualities qualities which are harder to replicate than most software. Making a Crysis clone is much more difficult than making a FOSS feedreader, for example, and people are more willing to pay for good vidya than yet another small program.

>>14574270

Put your game on a store and leave the source code open for people willing to compile it themselves.

You're very focused on short-term profits over longevity. It's an awful mindset for an independent developer and it's also one of the big problems killing AAA vidya.

>>14574297

Yes, and Bjarne too. Fight me.

>>14574231

Friendly reminder your shitty language belongs in the garbage can.

>>14574306

I've never been in a fight so i have to decline your offer

>>14574309

Name a better language

>>14574332

>Name a better language

C

>>14574335

>didn't say assembly

into the trash you go

>>14574332

machine code

>>14574304

You're not understanding what I mean. When I talk about "open source" I am talking about what the "open source movement" advocates here: https://opensource.org/osd Which if you read it you'll understand why I don't agree with open-sourcing any code I write.

You can create proprietary software that anyone can download and use for free. Consider the license that the X-Ray game engine used for STALKER uses: https://raw.githubusercontent.com/avoitishin/xray-16/master/License.txt . This kind of license ensures that anyone can use and modify and redistribute the code for free, but if they want to make money off of the code, they have to pay the developers of that code for a commercial license.

>>14574335

And what part of >>14574231 does not work in C?

>>14574231

I don't understand what the point of a constant is. It's like a inserting a butt plug into your anus so you remember not to fuck yourself in the ass.

>>14574356

auto

>>14574356

auto, nullptr, reference, const after function parameters

>>14574356

The nullptr is not a C keyword, in C you would have to use NULL. C does not have the "auto" keyword and you have to declare "const" before the type, not after. You also cant write something like "const&" , "&" is not used that way in C. You also can't write "const" in between the function header foo() and the first bracket.

>>14574373

int const* p
int* const p

does work in c

>>14574389

You're a dumb nigger who can't read, congratulations. Delete your shame.

>>14574389

>you have to declare "const" before the type

why did you mean by this then?

>>14574369

Agreed. Someone please explain.

>>14574389

>>14574392

>>14574396

>>14574405

He's right, I was wrong. I hadn't seen anyone do it before so I thought it was some C++ feature.

>>14574355

Again, that model rarely results in profit for the code's author unless they have a large business actively backing and maintaining the codebase, such as Unreal. I can guarantee you no one will contact the skeleton of GSC and attempt licensing X-Ray for commercial use, while Godot is earning nearly $8k a month from developers willingly paying them to maintain and improve their shit. Maybe you'll figure out someday how attaching price tags != profit after you rub the skin off your hands.

>>14574415

If you don't want a variable to be modified then don't modify it. Why do you need a special variable in order to not do it?

>>14574370

>>14574371

>>14574373

Auto is a keyword in C, "auto const float" is a valid thing. Note the lack of a semicolon there.

nullptr can be defined manually. #define nullptr NULL

Can't say anything about the use of & and putting const after foo() since I don't use C myself.

>>14574369

>>14574405

Const is stored in the program instead of having to be initialized during runtime. Without using const, your program effectively has both a variable AND a const to store the value to set the variable to.

Think about it as const being stored in ROM, while a non-const variable is in RAM. Except not hardware wise, but in software.

>>14574369

Magic numbers are bad. Constants are there for if you need to use the same number in multiple places and change it easily.

They're also there so you don't have to wonder what the significance of the number 691488 is three years later.

>>14574444

checked

>>14574369

>>14574405

You say, "Why use a const variable? Why not just not change the variable?"

…that's entirely the point of using the const keyword. So you CAN'T change the variable. If you don't use const, you might end up changing the variable in your code at some point, having forgotten it shouldn't be edited. Using const prevents you from doing that. It's an added check, not a necessity.

>>14574396

What if I told you that the C syntax is fucking retarded around pointers, and that the name of the variable can be considered the type?

const int* p;
int const* p;

The above two are actually considered equivalent in C. >>14574373 is actually technically wrong.

But the thing that makes C pointers really fuck is this:

const int p1, *p2, p3, ***p4, * const p5;

p1 is a const int, p2 is a pointer to const int, p3 is a const int, p4 is a pointer to pointer to pointer to const int, and p5 is the only thing that's actually a const pointer to const int. Whether it's a pointer is part of it's "name", and so is the const.

That said, C's usage of const pointers IS actually different to C++'s! In fact:

const int* const * const p;

I'll stop posting about it now since this is just arguing about the stupid bullshit that is the C standard.

>>14574438

>Auto is a keyword in C, "auto const float" is a valid thing. Note the lack of a semicolon there.

Sorry, I forgot that the auto keyword is a valid thing in C. It's just that you never see it since it is completely pointless.

>nullptr can be defined manually. #define nullptr NULL

(int)nullptr is guaranteed to be 0, whilst (int)NULL is not. They are not equivalent. They are however guaranteed to evaluate equal to (int)0.

>>14574417

If nobody will use it commercially any way like you say, why do you care? I don't really expect anyone to use my code commercially- it's just in principle.

Why don't we call everyone else who charges money for their work a jew? But no, because it's software, it's somehow different. I'm sure GSC is full of jews too- they didn't release their software under a free license like the GPL!

>>14574444

Use defines to name magic numbers.

>>14574473

>medium sized or large project

Depends what that means, my largest project is around 9500 lines. I can't recall ever running into a situation where I accidentally modify a variable I didn't mean to. I wouldn't go and modify a variable if I didn't remember what it's purpose is.

It does make sense for other people working on the same code though, and I can't say I have any experience with that.

This thread has gotten real autistic real fast. Not that I'm complaining.

>>14574496

Nice strawman, faggot.

Why do I care? Longevity. It pains me to see projects rot away and developers getting their ego tied up with their code. There's loads of profitable free software out there but the way many game developers talk, you'd think free software is some nearly dead joke, everyone buys random programs off the web, and web hosts pay Microsoft for server software.

>>14574542

This is the false choice that software has to be free, or nobody will be able to download and work on the code. I've already made it clear that literally my only objection to FOSS is the commercial usage of my software, and that I agree with every other part of it, in my posts on this thread.

Game developers act this way because they don't hold FOSS to the same kind of reverence that people who are in the Linux community do. I've already explained the issues with FOSS, and how you don't need to release your code as FOSS to prevent your project from rotting away with time. The x-ray engine is proprietary, yet projects like this ensure that it won't just die once the original code stops working on windows 11 or whatever. Putting the engine under the GPL is not the only option.

https://github.com/OpenXRay/xray-16

https://github.com/OpenXRay/xray

>>14574610

That's true but it sure makes it easier, compare the prevalence of id software engine source ports to those of the xray engine.

>>14574540

Less than a third of the posts ITT thread are on topic

That's the other reason I killed my thread

>>14574610

I prefer permissive licenses like the MIT and the three/two-clause BSD licenses, although the GPL has its merits and is a better option if you want to release your source code and ensure anyone modifying your shit does the same.

>>14574653

I've been reading beaverposter screencaps and GDC slides instead of working today. Might also skim the source of some other RPG-focused engines like GemRB or EasyRPG to see how they handle persistence across stages.

>>14574702

Can you blame them? Working with the microsoft toolchain is pure suffering.

>>14572958

1st pic looks like he's Russian.

>>14574712

I am a huge windows fag, but I don't use visual studio or any Microsoft products to write my code because I know how lame they are.. . Really I think it's more that everyone learns how to program on either Windows or Linux, and then to them the other side is really painful to program on, if you're writing low level code like for a game engine. I think that when you write code on one platform, you eventually "get it" and everything makes a lot more sense. Then when you go to the other one, you have no context for why things are the way they are, and have a bad time writing for it.

>>14574738

more like, you start on windows then go to linux and never want to go back to windows

>>14574132

>>14574302

>world cooors and tri-planar

Understandable

As that anon said for certain steps you just need data, and in retrospect it would be best to use a compute buffer here (instead of a 1D Tex); which you can set on the CPU and pass to the GPU for usage.

As long as you just read from the buffer on the GPU you’ll be fine. Though as a side note for the buffer make sure the stride size fits the average GPUs cache line size (fewer cache misses, it can improve speed by quite a bit compared to naive implementations).

>>14574743

I have tried to write for linux, I don't like unix as much as I like windows. If you write something high level then it's all the same but I think that microsoft made a lot of improvements to the kinds of API's you have to use on linux, I don't know maybe I should learn XCB instead of Xlib and see if they did better than win32 but even though I've written some stuff on linux I just haven't gotten into the unix way of doing stuff , I just don't think it's as good . Probably a side-effect of what happens when you actually get good at writing win32 code and then try an d port your stuff to linux.

>>14574702

Nobody plays games on Linux.

>>14574770

That's because the only games that are acceptable for FOSS fans to play are arena shooters (what other GPL games are there?)

>>14574778

Shitty clones like Tux Kart and games like Wesnoth that are ruined by the developer going muh vision whenever someone complains about gameplay.

>>14574778

Tuxkart, a bunch of strategy games, computer versions of many board and card games, roguelikes, and a lot of other stuff. A large portion of it doesn't look too great, but often is fine gameplay wise and you're free to make it not shit yourself anyway.

https://libregamewiki.org/GPL_games

>>14574778

There’s always the spring engine derivative games. Some quality games due to the original idea being inspired by TA.

>TA is the best RTS

>>14574778

>>14574770

Linuxfags do play games and have loads of games, both native and emulated. Strict freetards have a more limited selection but their games are exactly the shit they're interested in (arena shooters, traditional rougelikes, strategy games, board game clients, a flight simulator, and some janky platformers and racing games) so they're more or less satisfied.

>>14572920

>that's where you're wrong, /v/ always made the best to stay away from /pol/ bullshit

Son, 8chan /v/ was born from "pol bullshit". Please leave redditfag. You are not welcome.

>>14572958

>>14574539

> Depends what that means, my largest project is around 9500 lines. I can't recall ever running into a situation where I accidentally modify a variable I didn't mean to. I wouldn't go and modify a variable if I didn't remember what it's purpose is.

9500 lines is trivial.

>>14570689

There's nothing wrong with being an autist with a dream.

>>14573159

From what i've gathered get_tree() is used to get the scene tree. along with get_root() which i believe gets the root node of the scene. The root node isn't the first node you chose to create your scene but the world (or viewport in godot's case?) itself. Run your game and see what's happening in the debugger, you'll see what i mean.

>>14575144

It's not obfuscation, it's a space-saving measure. Limiting the file to as few lines as possible saves on disk space.

>>14575144

If I recognize the specific example, it's so he could fit it on a business card

people either obfuscate for fun, job security, or because they honestly think harder to read code means they're smarter

>>14575153

> job security

those damn fuckers, trying to make themselves indispensable.

>>14575213

it's okay, most people don't know they're obfuscating their code

like that anon who hates consts

>>14574369

>>14574405

1. Saying that you don't need const, because you can just not change the variable is saying you don't make any mistakes while coding. The keyword ensures you don't make this particular mistake. It becomes particularly important when dealing with classes in C++, because there you have to deal with recursion. If you declare a method of your class as const, then three things happen:

a) the compiler won't let it change anything about the object

b) the compiler won't let it call any method that isn't also const itself (thus ensuring that the object doesn't change recursively)

c) if you have a const pointer to an instance, the compiler will only let you call const methods

2. C and C++ weren't made for one-man armies of amateur game devs who think they can just write bug-free code and that therefore there won't be any problems. Just because you know that a variable isn't supposed to change, doesn't mean that everyone who sees your code also does. By using const you make that clear in an unambiguous way.

3. C and C++ are compiled languages and two things you want to do are static analysis and compile-time optimization. The effectiveness of both hinge on how well your static analyzer and compiler understand your program. The problem is that they are really stupid, so you need to make your code as expressive as possible for them to work properly. The const keyword isn't just there for you, it's also the tool with which you communicate the fact that something is a constant to, say a compiler, because it otherwise might not pick up on that fact.

There are optimizations that your compiler can only perform, if it knows that it's dealing with constants. Optimizing loops would be a simple example. Take a look at the following code:

Class *a = new Class();
for (int i = 0; i < 10; ++i)
{
    a->AnyMethod(i);
}


Class *b = new Class();
for (int i = 0; i < b->Count; ++i)
{
    b->NormalSomething(i);
}

const Class *c = new Class();
for (int i = 0; i < c->Count; ++i)
{
    c->ConstSomething(i);
}

In the first example, we're dealing with a compile-time constant. So what the compiler is going to do is called "loop unrolling". It will just get rid of all the loop logic altogether and insert ten calls to AnyMethod(). No counting is done, no conditions checked.

In the second example, the compiler can't do anything. Not knowing whether the call to NormalSomething() changes Count, it has to perform the loop in its unoptimized form. It will recheck the condition before each iteration.

In the third example, you're making c a const pointer. So the compiler won't let you change anything about the object through c. This means that ConstSomething() must be a const function, which also means that the loop body doesn't change c. Therefore Count is a runtime constant and the compiler can optimize the loop. It can't fully unroll it, but it will check Count once at the start and then execute ConstSomething() however many times it needs to.

>>14575088

This is so ugly, wasteful, and totally pointless I don't know why anyone would think it's a good idea

>Function called sfColor_fromRGBA needs a comment saying it constructs a color from RGBA

>Each of four 8 bit uints needs a line to say that they are 8 bit uints

>The return variable needs a comment saying it is being returned and that it is the result of the inputs inputs

Because that's literally all the comment blocks say.

Also, is this SFML C bindings? Why would you do that to yourself?

>>14575850

The purpose here is that these comments are used to generate web documentation with Doxygen. Granted, the style is fucking ugly. Sane developers use the standard Doxygen multi-line comment beginning with /** rather than the horrid triple-slash everywhere.

>>14575986

That's really cool, do you have the hardware to flash it?

>>14575986

You don't need to worry about the bankswitching bullshit, because in all the popular mappers, the memory is segmented into several "pages" you can swap freely, so you can use it to always keep your code in the memory.

Actually, in MMC1/MMC3 i think page 0 is always locked.

http://duramecho.com/ComputerInformation/WhatIsObjectOrientedProgramming.html

"The following article was written in year 2000 and is for readers who are used to programming but not to OOP. It explains the basics of OOP from a traditional (procedural) programming perspective and without the distracting hype that OOP was being evangelised with at the time I was writting. If you are a recently trained programmers whose first language was an object oriented one like Java, and so naturally think in OOP ways, this article might seem strange or even derogatory to OOP. It might nevertheless be slightly amusing though :-) .)"

>>14576005

Nope. Developing it strictly as a ROM for now, but testing on an array of emulators to be sure it should work on actual hardware. I'll wait until I actually get somewhere worthwhile before buying the hardware (until I at least have proper level loading using metatiles or real compression, and enemies with some interaction). Not much point dumping money into it when all I have is a single screen that you can't do anything in except move and jump.

>>14576014

That's good to hear, thanks. I have been putting off researching mappers and bank switching and getting as far as I can on NROM for now. I want to stick with a relatively simple mapper, and likely a Nintendo one. I'd like something that eases development to some degree, but the most complex functionality I probably need is switchable CHR and PRG banks and switchable nametable mirroring. I also want to stay as simple as possible in case I actually finish this fucking thing and want to sell carts, in which case a simpler common mapper is usually cheaper (the ROM is going to be free regardless). MMC1 might be all I need.

>>14576086

The biggest pain in the ass you will have is "infinite scrolling", specially if you want a bidirectional one.

There's a reason why many NES games just stay with either just horizontal or just vertical scrolling, or do a "dual tall, infinite right" thing like Super mario 3.

>>14575298

Man, I didn't realize it went that deep. I'm dumb.

>>14576109

Yeah, don't need "infinite" scrolling, just sets of single-wide or single-tall vertical and horizontal rooms that interconnect. The rooms will be quite long or quite tall, depending on the area, though, so I can't just depend on the double-screen scrolling some earlier games do, like Super Dodgeball. I've worked out some of those kinks, already, though.

It's an exploration-based game. I'm intending something like Metroid or Lizard when it comes to the scrolling and exploration mechanics.

>>14576214

Then it will be a fun ride.

Going full blaster master on your first try is quite terrible.

>>14578021

Nice try (((Mark)))

>>14578110

>taking advantage of the trust of my users and threatening the integrity of my game

>>14578110

Personally I'm hiding easter sperm in your mom

>>14578024

Speak for yourself, Mark.

>>14578110

I had an egg object in my game and I was planning on having seasonal tilesets. Only problem is that Easter has a rolling date because follows the lunar calendar

>>14578191

Did you accidentally import a brick texture as a grunge map for your egg?

>>14578238

Anon, this is real life.

No, it's an issue with the way I made them. I cut out the head shape out of electrical tape, and since that's thin, I had to use several strips. The lines of paint that resemble brickwork are sections in between the tape.

>>14574810

I'm honestly wondering what the fuck you're even trying to do.

>The problem is, it interpolates everything

You know you can generate barycentric coordinates just fine through interpolation, right? Just add a vec3 output attribute to the vertex shader, set it to (1,0,0) for one vertex, (0,1,0) for the second, and (0,0,1) for the third. The values will be interpolated, and the vec3 attribute that the fragment shader will be a barycentric coordinate for that fragment.

>>14577757

So sexy

>>14573385

>design document

isn't it equivalent to just having a project in github (or gitlab)?

the issue tracker is pretty good track feature changes and bugs, specially if you think of something good but can't do it atm

milestones are great planning up features and goals, for uh, milestones versions

and the wiki servers as a good meta documenting for the code itself, and some neat features like graphs and whatnot with their flavored markdown

>>14578796

>isn't it equivalent to just having a project in github (or gitlab)?

No, not at all, what do you think a design document is?

>>14578898

> what do you think a design document is?

a document documenting your game design

foremost a written description of what your game will be, what features and etc

and within the lines, the programming design you will be using to make things work

but they all are subject to dramatic change, so its a good a idea work on all things in parallel, documentation and hands on at the same time

thus, those gitlab things are rather useful helpers

>>14579147

They're "useful helpers" but that doesn't make them anywhere near equivalent. I have a feeling you're just writing whatever comes to mind without actually considering if it makes any sense.

>>14579368

Holy shit where did you come from

>>14579368

Much like any document about a project, it includes the scope of the project, lifetime of the project, requirements and other gay shit. It's only important if you're working in a team and not necessary.

>>14579368

would you fuck off back to reddit, rustfag?

So using version control with your project is pretty cool, it's a nice moral booster when your able to see how for you've come with your project looking back at all the commits you've made.

Why do these threads die at bump limit? This thread hit 250 posts in about 24 hours and since then it's been at a crawl, do people not post if a thread's not fresh?

>>14580521

I'm guessing people do this weird thing where they fall unconscious and their bodies take on this state of paralysis.

What do they call it again? Sleep or something?

>>14580534

I've heard about this. They also experience vivid hallucinations during the process.

>>14580534

>implying all anon's are in relatively close timezones

>implying anon's have a steady sleep schedule

I mean, I guess that's a possible reason, but this happens to these threads regardless of the time of day.

>>14580555

Checkem niggers.

I was trying to think up a floating point struct that had determinate decimal accuracy. For 32 bits of data:

>16 bit signed short (+/- 32767), acts as a multiplier

>8 bit unsigned byte (0-255), acts as an accumulator

>8 bit unsigned byte (0-255), acts as a fractional byte over 255

The number range is +/- 8,388,864, plus a fractional part of N/255. Is this a feasible idea? It mostly would just be for storing a value rather than evaluating it. Alternatively, I can change it so that the sign is actually an sbyte fraction so that 2 digits of accuracy is guaranteed

>>14580575

>Why would you do it when the thread is about to fall off the catalogue and gain no exposure, when in a few hours you'll be one of the first posts?

I kind of understand that, I guess it's just a flaw of how these threads work. A cyclical thread could fix this, but I remember someone saying that it makes archival harder.

>>14580608

>>14580575

We could solve that… but it would have a dependency on the thread OP.

After we reach the bump limit, and discussion has died down; the thread OP can archive the thread, delete the thread, and then make a new one.

We would lose the consistency of threads being interlinked, but we could replace that with a link on the wiki; like updating a "current thread" link on the homepage.

May help revive the wiki too.

>>14574369

Compiler can optimize the code better if you are clear about the use and intentions of your variables and code. Which leads to, better error messages and error checking. If the compiler knows a function shouldn't be making changes through a reference or pointer (because you declared it as a const input variable), it can enforce this within the function.

const keyword can head off a lot of typical issues with pointers and references. It can also make your code faster. I mean it's already C#, Java, about 1 billion times faster than Python, but yeah. Even faster.

Oh wait, I'm retired, it's a different anon ban both of them

>>14580868

I would like to think that we only have one reddit nigger here, please don't make me paranoid about it

>>14580922

That post you capped is also me, telling rotten to go away in a thread from a while ago. Look at the id's, the post with tomoko_slut.png is me, and the other two are him responding to me confirming that he is the one who has been derailing the threads with his bullshit and confirming that he is from reddit.

>>14580948

>>14580924

Yeah, I realized too late I am actually retarded

How do we get mods to stop him from derailing the threads.

>>14580963

Not much that can be done. Someone can't broadcast "I am from reddit, I am currently crossposting, look at this cool reddit post I found" and not get shit on.

>>14580967

A permaban is definitely warranted because of the constant derailment and ruining of these threads. In a vacuum these incidents that should get you a temp ban. But when you keep coming back, and keep doing the same thing, and keep ruining the general, it makes me wonder why the moderators haven't done it yet.

new thread soon

hopefully whoever makes it doesn't have shit taste like the last guy

>>14581116

Good enough taste?

>>14581585

>>14581585

>>14581585

>>14581585

>>14581585

>>14581585

NEW BREAD

>>14579441

>Much like any document about a project, it includes the scope of the project, lifetime of the project, requirements and other gay shit. It's only important if you're working in a team and not necessary.

and that's exactly what I've said before…

>>14580521

>Why do these threads die at bump limit?

guess what, because most people here are actually shitposter immigrants while, real developers only lurk and post useful things every once in a while,

specially if they want to make the best of their time developing….

and yes, version managing is pretty good