What are you working on?

[proboardslol]

[QUOTE=roastchicken;49682259]My shitty terraria clone now has textures!

[IMG]http://i.imgur.com/Rtr6rPH.png[/IMG]

having two papers due really gives you the motivation to work on anything else[/QUOTE]

I know what you mean dude. Once, I had to write a half page paper on Anne bradstreet, so I developed an AI and passed a turing test, thereby revolutionizing computer science.

[Socram]

[QUOTE=Cyberuben;49681150]Due to a lot of articles I read lately, I'm completely done with wanting to work in the game industry. I can't be motivated the slightest to work on games. I'm currently enrolled in a game development study and have been for the last two years (right now halfway the 3rd year out of total 4), but I'm not looking forward to my internship or working at AAA companies but indie doesn't interest me either.

I think general software development is something way more fit for me, so I'm probably going to get my diploma in general ICT instead of game development.

What interests you guys most as gamedevelopers?[/QUOTE]

Don't ask us to convince you to stay interested in game development, if you're already burnt out on it then it sounds like maybe it wasn't a good fit in the first place. You shouldn't base your life decisions on game dev articles on the internet, but you also shouldn't base it on our personal opinions and experiences. Be your own person and just do what you want to do, your degree is largely irrelevant. If you "can't be motivated in the slightest to work on games" its not the articles fault or anyone else, it just sounds like your not interested in it. Do what you say you're interested in.

I love being in game development, and so far my professional career as a game developer has been rad as hell.

[DrLuke]

[QUOTE=paindoc;49679462]What would be the best way to visualize or plot, preferably in 3d, the output of some C code I'm working on? I'm just converting my matlab octree and n-body stuff to C, and while speed improvement is nice and not using MATLAB is also sorta nice (:v:), the ease with which I was able to visualize and test the output of my code is going to be missed.[/QUOTE]

If you're willing to try linux you could try installing the linuxcnc-sim package and basically get a full g-code interpreter with a simulated cnc machine:

[t]https://my.mixtape.moe/ywsmgi.png[/t]

Another very good option is to use openSCAM: [url]http://camotics.org/[/url] (Also available for windows)
In there you can define your tools and it will render you a 3d model out of your gcode. Neat!

[jmazouri]

I was trying to see if I could get WASAPI loopback capture working in Unity and then this happened

[vid]https://my.mixtape.moe/fukzfm.webm[/vid]

[NeverGoWest]

[QUOTE=Cyberuben;49681150]What interests you guys most as gamedevelopers?[/QUOTE]

Some people tell me that I never lost my childlike fantasy and creativity, which is very useful when making media. I have aspired to be a load of things, horror author, play in a band, even wanted to make short movies or an animated series. Then I realized that I have never ever played a game without feeling it lacked something or could do well without. So that is my motivation: To make games that don't suck in my own opinion, to express my creativity, to not be a shitty developer or publisher instead of microtransactions, DLC and paying for ads, I would rather have word of mouth be my market model. If people think my games are great and worth the money, it will get out there.

A great game doesn't need voices or some deep and intricate story. A basic premise well executed can make history.

This is what motivates me.

[Berkin]

[QUOTE=Cyberuben;49681150]What interests you guys most as gamedevelopers?[/QUOTE]

There's a whole fucking world of weird shit up here in my head and I have all the necessary tools at my disposal to realize it in the physical realm. Such an ability is both wonderful and empowering to me, especially considering how far graphics and computing technology has come in the past decade alone. The experience of creating an entire universe with whatever you want in it is what keeps me pursuing this career. There is no limit to what can be made.

We aren't limited to just writing books anymore, where everyone might interpret and visualize the world you've created differently. In video games, you can tell a story where everyone gets the same experience that you first conceived in your mind's eye. Not only that, but the story can adapt and change, and players can interact with its elements as if they were actually there. Isn't that goddamn amazing?

[Richy13]

[QUOTE=adnzzzzZ;49680754]wrote a blog post on why i think copypasting is good [url]https://github.com/adonaac/blog/issues/10[/url][/QUOTE]

Why not have:

function _1()
  block 1
end

function _2()
  block 2
end

function _3()
  block 3
end

function _9()
  block 9
end

function _10()
  block 10
end

function A()
  _1()
  _2()
  _3()
end

function D()
  _1()
  _9()
  _3()
  _10()
end

function X()
  A()
  B()
  C()
end

function Y()
  D()
  B()
  C()
end

That way your code isnt duplicated but you can still achieve the structure you want without ugly flags and branching (I dont know about lua/luaJit but with most compiled languages you can instruct the compiler to substitute the function call with the code)

Of course all of this is just assuming that the block # code is a few lines or more, if its just 1 line of code then there is no point in doing this

[mastersrp]

[QUOTE=adnzzzzZ;49680754]wrote a blog post on why i think copypasting is good [url]https://github.com/adonaac/blog/issues/10[/url][/QUOTE]

You really should give [url=http://fossil-scm.org/]fossil[/url] a look. The whole github issues thing is weird.

[backwardspy]

[QUOTE=Sidneys1;49681921]Are you on build >1511 and using Command Prompt (and not Powershell)?[/QUOTE]
I am using command prompt, but I have no idea which build I'm on. I checked using 'winver' and it says build 10240, which doesn't fit the YYMM format that it's apparently supposed to be in?

[cartman300]

[QUOTE=BackwardSpy;49683134]I am using command prompt, but I have no idea which build I'm on. I checked using 'winver' and it says build 10240, which doesn't fit the YYMM format that it's apparently supposed to be in?[/QUOTE]

You have to disable legacy console.

[thumb]http://i.imgur.com/wKjM8k8.png[/thumb]

[Silentfood]

Now that I've finally recovered, my [url=http://nv-mp.com]NV:MP project[/url] went from ~100 signups to 10,000~ in 48 hours.

I started rewriting the network framework, haven't implemented ENET yet but I'm going to try and make this more organised so I can deploy this to other projects w/o much hacking.

[url]https://github.com/Silentfood/NETTIK_Networking[/url]

[backwardspy]

[QUOTE=cartman300;49683183]You have to disable legacy console.

[thumb]http://i.imgur.com/wKjM8k8.png[/thumb][/QUOTE]
Already did that a while ago.
[t]http://vgy.me/qakOEC.png[/t]

[Drury]

[QUOTE=Cyberuben;49681150]Due to a lot of articles I read lately, I'm completely done with wanting to work in the game industry. I can't be motivated the slightest to work on games. I'm currently enrolled in a game development study and have been for the last two years (right now halfway the 3rd year out of total 4), but I'm not looking forward to my internship or working at AAA companies but indie doesn't interest me either.

I think general software development is something way more fit for me, so I'm probably going to get my diploma in general ICT instead of game development.

What interests you guys most as gamedevelopers?[/QUOTE]

Tell me the name of 1 good Slovak videogame.

I want to make at least one before I die.

[adnzzzzZ]

[QUOTE=Richy19;49683016]Why not have:[/QUOTE]

This has the exact same problems as the ones I outlined in the article. If you want to add a new feature Z that uses a block of code very similar to the one inside _1 but kind of different, you'll have to go into the _1 function and either add flags or make a new function that is similar to that block of code but with the differences that comply with Z's needs. I could have stated in the article that each block is also subject to change right, because that's something that happens too. I didn't because I didn't want to go to deep in the details of this particular example since it's just there to make a point.

[sidneys1]

[QUOTE=BackwardSpy;49683134]I am using command prompt, but I have no idea which build I'm on. I checked using 'winver' and it says build 10240, which doesn't fit the YYMM format that it's apparently supposed to be in?[/QUOTE]

[t]http://i.imgur.com/QJqDDDL.png[/t]

EDIT: Or use 'ver' in command prompt. Mine prints:

Microsoft Windows [Version 10.0.14251]

EDIT AGAIN:
Regardless, 10240 is the RTM. (Hence the multiple of 1024) You need to be on the insider track to get these newer builds. Newest public was the Fall Update which (iirc) was 10586, while I'm about to install 14257 as an 'Insider Fast' member. I [URL="https://buildfeed.net/source/PublicRelease/"]get new builds [/URL]every two or three weeks. [URL="https://buildfeed.net/"]Internally, it's even more convoluted.[/URL]

[Silentfood]

[QUOTE=adnzzzzZ;49683478]This has the exact same problems as the ones I outlined in the article. If you want to add a new feature Z that uses a block of code very similar to the one inside _1 but kind of different, you'll have to go into the _1 function and either add flags or make a new function that is similar to that block of code but with the differences that comply with Z's needs. I could have stated in the article that each block is also subject to change right, because that's something that happens too. I didn't because I didn't want to go to deep in the details of this particular example since it's just there to make a point.[/QUOTE]

About the whole "copy - paste" being right or wrong, it all depends on context. If you're prototyping, it's completely acceptable to copy-paste and hack your way through. You're expecting some output, so as long as you're getting it you're fine. But if for example you're working on a "release" of your program, or approaching a milestone you really should be using good coding practices that normalises your code appropriately.

You make it sound like using object inheritance and breaking down routines are chores and take too much mental strain, but if you're already confident in these practices it shouldn't be such a task. I'm also pretty sure that the OOP model is designed to stop code reuse, not support it.

Just my two cents on your post.

[adnzzzzZ]

[QUOTE=Silentfood;49683641]But if for example you're working on a "release" of your program, or approaching a milestone you really should be using good coding practices that normalises your code appropriately.[/QUOTE]

I disagree. The whole point is that it's a tradeoff and when you know what you get and what you lose you can make these decisions even in the final version of your code.

[QUOTE=Silentfood;49683641]You make it sound like using object inheritance and breaking down routines are chores and take too much mental strain, but if you're already confident in these practises it shouldn't be such a task.[/QUOTE]

Even if you're very confident on them they are by definition more mentally straining than just copy pasting. This isn't my opinion. It takes by definition more mental resources to think about making your code properly reusable than just copy pasting. And the main problem is that this is a O(n) mental cost, where n is the number of components/classes/methods/mixins that are relevant to whatever functionality you want to change/use/add. Whenever you add some functionality, now you have to think about how ALL the abstractions you made fit with the new thing you have to add and which ones you're going to use or change. You're adding new functionality AGAINST existing code. While with copy pasting its O(1) because you don't have to think about anything.

[QUOTE=Silentfood;49683641]
I'm also pretty sure that the OOP model is designed to stop code reuse, not support it.[/QUOTE]
??? I'm pretty sure one of the big goals of OOP is increasing code reuse.

[Cyberuben]

[QUOTE=adnzzzzZ;49681788]Why not take your life into your own hands and just start doing good work as some kind of freelancer? Why waste more time and money on a degree when you already know what you want to do (web development) and you already have the skills necessary to improve even more. Why do you need a degree? And why do you have to rely on the industry of your country existing to make games? It's literally the best time right now to be an indie developer with all the data I have pointing to Steam being a super reliable income stream if your work is above a certain level of quality. Again, why not take your life into your own hands and improve it yourself instead of relying on other people (the industry)?[/QUOTE]

I am already a freelancer, have had my business for 3 years and I do make some good money on the side, however, I'm only 20 and I really want a degree because I see where my parents are now, after the economical crisis, and they both wished they had gotten a degree when they were young. A diploma alone doesn't guarantee you a job, and I'm convinced that a portfolio is more impressive than a piece of paper that says "you did well in school", but a combination of both will definitely look good and I still think that a university can teach me some things, so that I can explain myself to the client better, write proper project plans before starting to work on things and hoping it'll just work in the end, and so on.

Wanting a degree is partially future-proofing and partially thinking I can genuinely learn things there.

And my motivation for not going indie: I'm everything but creative when it comes to coming up with game ideas, I can't do art and I do not have the funds to start hiring people for this.

Programming wise, there are too many aspects of game development that I find tedious and do not want to develop myself further.

It's just a bit shit that I had been looking forward to studying to become a game programmer for 3 years, and now after 2.5 years of actually studying it I'm already done with it. Better late than never, I guess?

[Silentfood]

[QUOTE=adnzzzzZ;49683673]I disagree. The whole point is that it's a tradeoff and when you know what you get and what you lose you can make these decisions even in the final version of your code.

Even if you're very confident on them they are by definition more mentally straining than just copy pasting. This isn't my opinion. It takes by definition more mental resources to think about making your code properly reusable than just copy pasting. And the main problem is that this is a O(n) mental cost, where n is the number of components/classes/methods/mixins that are relevant to whatever functionality you want to change/use/add. Whenever you add some functionality, now you have to think about how ALL the abstractions you made fit with the new thing you have to add and which ones you're going to use or change. You're adding new functionality AGAINST existing code. While with copy pasting its O(1) because you don't have to think about anything.

??? I'm pretty sure one of the big goals of OOP is increasing code reuse.[/QUOTE]

I misphrased the last part, but my point still stands: why would you copy paste code in an OOP environment? I think you're too focused on what works for you. You wouldn't get away with doing them types of practices at a software house or in a team based project.

You're flawed in understanding how software is valued. How would you sell that code to a client? Do you really think that your employer would be happy with this code practice? If you were a team leader, do you think that your team members would be happy with this approach?

[Drury]

[QUOTE=Cyberuben;49683692]I am already a freelancer, have had my business for 3 years and I do make some good money on the side, however, I'm only 20 and I really want a degree because I see where my parents are now, after the economical crisis, and they both wished they had gotten a degree when they were young. A diploma alone doesn't guarantee you a job, and I'm convinced that a portfolio is more impressive than a piece of paper that says "you did well in school", but a combination of both will definitely look good and I still think that a university can teach me some things, so that I can explain myself to the client better, write proper project plans before starting to work on things and hoping it'll just work in the end, and so on.

Wanting a degree is partially future-proofing and partially thinking I can genuinely learn things there.

And my motivation for not going indie: I'm everything but creative when it comes to coming up with game ideas, I can't do art and I do not have the funds to start hiring people for this.

Programming wise, there are too many aspects of game development that I find tedious and do not want to develop myself further.

It's just a bit shit that I had been looking forward to studying to become a game programmer for 3 years, and now after 2.5 years of actually studying it I'm already done with it. Better late than never, I guess?[/QUOTE]

Not meaning to come off as an expert on your life but you sound to me like someone who thinks they can do everything by themselves and then naturally can't get motivated to do anything.

I may be wrong, but if that's the case and you're finding yourself throwing the towel, you're probably not fit to live that kind of life whatever you decide to do.

[adnzzzzZ]

[QUOTE=Silentfood;49683839]I misphrased the last part, but my point still stands: why would you copy paste code in an OOP environment? I think you're too focused on what works for you. You wouldn't get away with doing them types of practices at a software house or in a team based project.[/QUOTE]

Well yes, I prefaced the article by saying how all this is especially true if you're an indie developer working on an indie game. Of course most of this isn't viable for a team of multiple people. Although the number of people isn't the most relevant metric here, but the longevity of whatever you're building. For games this longevity is low. You usually build the thing, ship it, update it a few times and that's done and you move on the next project. This cycle takes anywhere between 6 months and 2 years, but overall this is a small amount of time, so small that the long term costs of copypasting often won't come into play as much as you'd think.

[QUOTE=Silentfood;49683839]You're flawed in understanding how software is valued. How would you sell that code to a client? Do you really think that your employer would be happy with this code practice? If you were a team leader, do you think that your team members would be happy with this approach?[/QUOTE]

I think you're misunderstanding my point. My point isn't that you should copypaste all the time. It's that you should look at more than code reuse as things that you want to optimize for. And one of the things that comes out when you do that is that code duplication isn't that bad of a thing to have in your code base under a lot of circumstances.

But to answer your question, clients don't care as long as it works. I'm not doing this in a dramatic way in other people's codebases. And if I were a team leader I'd definitely be worried about people who are too eager to abstract things too fast, as in my experience this leads to more complicated codebases.

[Fuyukai]

What Are You Working On - Somebody gets into an argument with adnzzzzZ [B][I]again[/I][/B] edition

[Richy13]

[QUOTE=adnzzzzZ;49684013]Well yes, I prefaced the article by saying how all this is especially true if you're an indie developer working on an indie game. Of course most of this isn't viable for a team of multiple people. Although the number of people isn't the most relevant metric here, but the longevity of whatever you're building. [B]For games this longevity is low. You usually build the thing, ship it, update it a few times and that's done and you move on the next project.[/B] This cycle takes anywhere between 6 months and 2 years, but overall this is a small amount of time, so small that the long term costs of copypasting often won't come into play as much as you'd think.[/QUOTE]

Maybe a few years back, but that really isnt true these days. It all depends on what kind of game you are working on, and ofcourse for pc/indie stuff it is still mostly true, but for mobile for example there are games that came out years ago and are still going strong getting big updates regularly

[QUOTE=adnzzzzZ;49684013]I think you're misunderstanding my point. My point isn't that you should copypaste all the time. It's that you should look at more than code reuse as things that you want to optimize for. And one of the things that comes out when you do that is that code duplication isn't that bad of a thing to have in your code base under a lot of circumstances.[/QUOTE]

Sure, but on the other hand, what if you have some functionality and all your game objects have this functionality and for some reason or another rather than making them inherit it you have copy pasted this code.. Now you want to make a fundamental change to how it works, you have to edit that code everywhere its been copied to

[QUOTE=adnzzzzZ;49684013]But to answer your question, [B]clients don't care as long as it works[/B]. I'm not doing this in a dramatic way in other people's codebases. And if I were a team leader I'd definitely be worried about people who are too eager to abstract things too fast, as in my experience this leads to more complicated codebases.[/QUOTE]

I think you are confusing clients with consumers.. If I sell a product to john doe down the street he just wants it to work, but if a company hired me to deliver them some piece of code they would want it to be written in a way its easy to work on if they then need to change something.

Honestly a good analogy of all of this, are databases. This whole thing is why database normalisation exists

[ichiman94]

January feels like the longest month ever. It doesn't help that next semester starts at the second next monday.

[AwesomeX]

If no one else is going to be creating a thread for February. I could, it'd just take some time to look through this one for highlights.

Should I?

[adnzzzzZ]

[QUOTE=Richy19;49684099]
Sure, but on the other hand, what if you have some functionality and all your game objects have this functionality and for some reason or another rather than making them inherit it you have copy pasted this code.. Now you want to make a fundamental change to how it works, you have to edit that code everywhere its been copied to[/QUOTE]

You're simplifying the situation way too heavily here. I already spent an entire article talking about this. It's all about tradeoffs. If you're using something on literally all your objects it would be stupid to copy paste it everywhere. I don't know why you assume this is what I'm saying.

[profan]

This is also an interesting read on the same subject: [url]http://www.sandimetz.com/blog/2016/1/20/the-wrong-abstraction[/url]

[cartman300]

You guys argue about stuff, yet have exactly the same stance on everything.

Guy A: It's easier to copypaste code than to abstract it
Guy B: It's nicer to abstract it than to copypaste it everywhere

Yet both Guy A and Guy B don't actually want to copypaste the same code a thousand times or make a God method that handles every use case ever.

Your arguing is pointless. You all say the same shit and then argue about who had the better wording.

[editline]6th February 2016[/editline]

It's like arguing against yourself.

[Tamschi]

[QUOTE=Cyberuben;49681150][...]

What interests you guys most as gamedevelopers?[/QUOTE]

Communication I think. I know that's vague and unspecific, but that's what it boils down to for me.
The programming puzzles are a nice bonus though, if they're interesting.

[editline]6th February 2016[/editline]

[QUOTE=Sidneys1;49681676][...]

As a bonus, if you use the 'type' command with [URL="http://artscene.textfiles.com/vt100/"]old *.vt files[/URL]... We'll, I'll let you find out for yourself. :)[/QUOTE]

It supports (most of) the control characters but unfortunately doesn't pause for me, so the animations go way too fast.

[editline]6th February 2016[/editline]

[QUOTE=proboardslol;49681975][...] the webm host reuses link and deletes old videos [...][/QUOTE]

Which host?

[editline]6th February 2016[/editline]

[QUOTE=adnzzzzZ;49683673][...]

Even if you're very confident on them they are by definition more mentally straining than just copy pasting. This isn't my opinion. It takes by definition more mental resources to think about making your code properly reusable than just copy pasting. And the main problem is that this is a O(n) mental cost, where n is the number of components/classes/methods/mixins that are relevant to whatever functionality you want to change/use/add. Whenever you add some functionality, now you have to think about how ALL the abstractions you made fit with the new thing you have to add and which ones you're going to use or change. You're adding new functionality AGAINST existing code. While with copy pasting its O(1) because you don't have to think about anything.[/QUOTE]

There's a limit for most people (in terms of single function length) from where they'll work with it more efficiently by reducing the visible set through encapsulation.

[QUOTE]??? I'm pretty sure one of the big goals of OOP is increasing code reuse.[/QUOTE]

No, not instantly. In fact it initially behaves almost exactly like procedural or functionally composited code in that aspect, assuming each is done well and you have method delegates.
What it really does simplify at a surface level is ([I]statically verifiable[/I]) copy-pasting, since you have a bunch of copy-pasted behaviours grouped in a class and they're selected implicitly when you use the instance.

I do almost exactly what you propose in your blog post for a current project of mine, letting the vtables handle function call routing.
The derived classes then only contain the data-format specific code that processes the instance's data in a way that's compatible to the rest of the program (if there's any local data at all. The class could just be a collection of behaviour validated against an interface).

Where OOP does deduplicate code (very efficiently) is if you have one behaviour collection copy and then want to make many (e.g. four or so) small behaviour variations all in the same position (and know the rest won't differ).
You can then simply implement everything else in an abstract class and fill in the rest in in a few derived classes. (It doesn't work if the variation locations differ; then composition or just plain class copy-pasting are far superior.)

Of course you can use these patterns in good functional or procedural languages as well, but generally speaking they require more boilerplate code and have worse error messages once you have these different similarity levels.
(You'd probably effectively implement vtables in those cases, since that's the most straightforward way of solving the problem without compromising your ability to quickly fix bugs.)

[polkm]

added fake 3d to the swords by using voxels, now they can spin :science101:

[IMG]http://i.imgur.com/RFALZ05.gif[/IMG]

[editline].[/editline]

a slower close up
[IMG]http://i.imgur.com/YvIMh0A.gif[/IMG]