Peer Pressure

General Discussions
#1

So I’ve got this friend who absolutely hates Java. Who can blame him, many people do. However, his reasons are completely unjustified. He attends the same college as me, and is currently working towards the same major as I am: Management of Computer Systems. Unfortunatly for him, our college teaches Java.

He’s a hardcore gamer (so much to the point where it effects his grades) and he’d much rather be learning C++. This is understandable since when was the last time you saw a commercial Java game on the shelves? Although I’d rather be learning C as well, I see the value in learning Java for two reasons: 1) Undergrad programming courses teach concepts and it’s much easier to grasp them in Java’s environment. 2) Java is growing and C is (debatably) dying. He cannot look past Java’s faults and C++'s history to understand that he must tough it out and just start with the coding.

He would love more than anything to begin development on a game- if only it were written in C. But what he doesn’t understand is that games are more than just syntax and tricks (he thinks he can learn by example from previous C++ games). It is difficult to make a complex working game- even a ‘simple’ shooter requires high level programming techniques. While I sit here getting down and dirty with such techniques, he’s muttering around in his dorm complaining about how C++ is so much better than Java. By the time he gets around to tackling C (with no professional instruction), I’ll have completed my Java gaming projects. By the time he gets through the harsher learning curve of C++, I’ll have started working on C myself. By the time he begins to map out programming techinques for his game, I’ll have already written them as I have done so before in Java and understand what’s needed.

How can I get him to understand the purpose of learning Java? How can I get him to break free from his hold on C++? He thinks things are so easy but I fear he’ll learn he’s wrong too late in the game. He needs to see the value in Java programming now while he’s still young and has access to professional instruction (he may not last much longer here with his additude). There comes a point where you have to stop trying to reinvent the wheel. Such is why I’m developing my first game in Java. Rather than read 20 different APIs for C++ and fail at getting Direct X initialized (without using someone else’s code), I’m writing from scratch, a full Java game. He’s spinning his wheels in C++. He needs to go forward. How can I make him understand that the only way he will go forward is with Java? I beg of you folks: Help me help him.

Note: I began my programming carrer in C++. I tested out of the classes he’s in now because of my past C++ experience so I definatly understand the boat he’s in. Compared to what I’ve seen in C, he’s only touched the tip of the iceberg. Trust me when I say he needs Java to move forward- he really will spin wheels with C++. I just don’t know how to make him realize it.

#2

With due respect, there’s nothing wrong with wanting to do game development in C++. It makes sense at the moment.

The only thing I can say is that it makes little difference what language you use, the actual concept looks pretty similar. If you’re writing good C++ you’ll find it looks alot like Java anyway, but of course most C++ developers don’t actually write good C++, they write C in C++ constructs. (personal observation)

Most important thing to note, most game developers write in C++, not because they don’t want to use Java, but that they either fear the change or simply don’t realise its possible. In addition C++ has a lot more support for games development than Java. However, unless your friend has huge wodges of cash to invest C++ will give him little or no support since all the games technologies cost through the nose.

In short, if he wants to use C++, go for it. When he starts seeing the difference between the difference between productivity between yourself and him it should start seeping in. If it doesn’t then he’s already too close minded to consider other possibilities, at which point its simply not worth the time.

Incidently, I started in C, did some C++, moved into Java for a few years, now I’m doing C++.

EDIT: Incidently, there’s no point using anything you get from this board to try and convince him, why would he take any notice of people who use the very thing he hates.

Kev

#3

What he really needs is to be forced to maintain someone else’s C/C++ source code. Or go work at a real company on unpaid work-experience - and be asked to track down a memory leak that is preventing them from shipping. That usually scares the pants off people sufficiently to make them run screaming back to java.

Or, you can point out to him that he is in good company: his reaction to java is one the games industry went through about 5-10 years ago, almost exactly the same as he is saying now. Only, it was with C programmers refusing to pick up C++, or refusing to use its OO because “it’s slow, and it makes my head hurt, and I don’t need it - I’m a genius programmer, I don’t need a crutch”. That lot soon lost their jobs and their careers, and the ones who embraced the new tech got all the money and worked on the successful games - which were now being produced to a higher level of quality in slightly less time, with more features (sound familiar?). Gross generalization, of course.

Basically, I would tell him he’s an idiot, to his face. Because that’s what he’s doing - turning himself into an idiot - by refusing to wake up and smell the roses. If he goes through life refusing to learn, and never working out the truth for himself, he will ALWAYS be a 3rd-rate programmer, he will NEVER get to work on great games (great games are nearly always written by great people…who can spot dead-weight a mile off) - the best he can hope for is to get away with being on a team with some good people, some great, and some other useless fools like himself, and just cross his fingers hoping the game will get to market before the company gets split up.

#4

[quote]What he really needs is to be forced to maintain someone else’s C/C++ source code. Or go work at a real company on unpaid work-experience - and be asked to track down a memory leak that is preventing them from shipping. That usually scares the pants off people sufficiently to make them run screaming back to java.
[/quote]
Haha! Exactly my point! It’s this sort of overhead he hasn’t even seen yet. He thinks C++ is simple, yet, he’s never heard of a pointer.

Kevglass:
I agree, there is nothing wrong with developing in C++. He argues the same: It makes no difference what language you use, so why can’t I use C? However, while he’s stuck on such schematics, I’m trying to ge thim to learn the concepts, which he’s not seeing either way. His C++ experience is very limited. Right now he’s putzing around with Ragnarok Online scripts for a server he’s a Gamemaster of. Before that he was looking into open source MUD development.

His goals are high, but the path to get there will be easier taken first in Java. For example, I recently posted about pre-loading strategies in Java 2-D game development. I’m actually going fourth with my game and tackling real problems faced by C++/Java programmers. I know what doublebuffering is, I setup sprite collision code (While rotating! It works! ), I’ve got great gameplay starting to develop for me. Sure he can tackle all those things in C++, but for the first run with things, It’s much easier to do so in Java than to run around following a treasure map of pointers (Go here, ok, go here, now, gohere, ok, there’s your actual data!). By the time he even figures out pointers, I’ll be well along my way of setting up the infrastructure of my first C++ game.

#5

I still think thats pretty closed minded. We’re not talking about rocket science here, writing simple C++ is pretty similar to writing simple Java. The chain of pointers is remarkably similar to the chain of objects in Java. Infact, pointers are pretty sickly in C++ nowadays, references are the order of the day where possible (again, pointers => C programmer trying to program in C++).

Now, please don’t get me wrong, I’m very pro Java, but for learning concepts the pros and cons are much more evenly weighted. If we’re talking about pure 2D, pick up any of the well written C++ 2D libraries. If we’re talking about OpenGL then the code is pretty much the same. If we’re talking about higher level 3D there are quality scenegraphics in both C++ and Java.

If we’re talking about full scale game development the trade offs in productivity are far higher. As blahblahblah says maintaining someone elses C++ is a nightmare. Not to mention the intricaces of different C++ compilers, memory model differences, bad “standard” library implementations not to mention random name mangling.

In short I don’t think you’re going to convince anyone by saying C++ is more complicated. Its almost an insult to some people, “you’re saying I’m not clever enough to understand it!”.

An interesting point I suppose would be given that all your game aspirations are working out, why do you care what route he’s taking?

Kev

#6

Actually, I suspect you’ll more likely be completing your first major java game. If you can code in Java then there’s (generally) little value in writing games in C++ (unless you’re forced to, e.g. by an employer).

Seriously, there’s only one thing that’s going to convince him: challeng him to a game programming competition. Both of you write games, and see who finished first, and what they produce.

If you don’t produce something both better and faster in less time then there’s either a serious imbalance in your relative abilities or something’s gone wrong. In which case, you’ll both learn a lot from working out what.

Present it to him as a “trial by fire”, an opportunity for both of you to get real factual metrics on comparitive ease and effectiveness of language. Whatever happens, you’ll learn something useful from pitting yourselves against each other, language to language, and if he’s half the gamer you describe, he shouldn’t be able to resist the lure of something seemingly “competitive”.

Oh, and even better: keep a diary as you go, weekly. It will make a fun and interesting case study for others to read :).

#7

Honestly, I care because he’s messing up his life. This battle goes beyond Java vs. C++. This battle is him vs. his “better things to do factor”. It’s trying to get him off his ass and put his many ideas to use. He almost failed out of college last semester for lack of attendence- he just played video games all day. Granted, right now he’s happy, however, I know he won’t be in 5-10 years (he says so himself). He wants to get out of college and program great games- he won’t do that with nothing to show of himself.

I’m the first person to advocate games over other things. I believe that competing in and learning a videogame is no different than learning to play a musical instrument (game dependant of course- barbie fun world is not a good example). They both have physical and mental benifits. Unfortunatly, right now, society doesn’t quite understand the good things games are able to provide. (www.sirlin.net for a more in-depth look at this) But this is besides the point- both things are wonderful as long as you get your P’s and Q’s in order. He needs to move forward with his life if he’s ever going to realize his dreams.

He already has a passion for making videogames. However, he’s doing nothing with it and feels badly about it. I figure that if I can instill some ambition within him, he’ll take initiative with his life. For the beginning programmer, Java is easier to see the light at the end of the tunnel. He’ll be able to work on conceptual problems in game development rather than tecnical ones with code syntax. I mean no disrespect to his ability to learn, however, I think C++ is too much for him at this stage of the game. There are many hoops and hurdles to jump through in game development- if you can focus on the ones geared towards gameplay itself, the game will be better in the end. He has huge ideas floating around in his head and I know they will not be realized in C++ simply because he gets too far ahead of himself.

#8

I highly recommend sirlin as “required reading” for people who want to be truly great games designers :). Especially the parts on scrubs. Read. Enjoy. Learn.

It might help to get him to read this:

http://stratics.com/content/articles/mmoguide.php

an article I wrote recently for Stratics Central (way big MMOG fan site). It addresses the same issue, but with an MMOG-specific slant. There are an awful lot of people like him who want to (and often try to) develop a “better-than-EQ” game; this article is for all of them :). When it was published I had nice comments on the article (short though it is) from some of the household names in the games industry, along the lines of “yes, this needs to be said”. Maybe he’ll think it’s cr*p; maybe it’ll make him sit up a little. Worth a try…

#9

I understand it’s sometimes messier to code in C++, but why would there be little value in coding with C++ (from my point of view, it’s only an issue of development time)?
Also, doesn’t low-level C++ stuff help in some cases?

#10

[quote] This battle is him vs. his “better things to do factor”. It’s trying to get him off his ass and put his many ideas to use. He almost failed out of college last semester for lack of attendence- he just played video games all day.
[/quote]
This guy sounds like a raving moron and lazy dullard. Tell him if he wants to learn c++ so badly then transfer to a school that teaches it, and pursue a computer science degree, not a business degree. (Here in the states thats what mis is, basically) Which begs the question; why is he even in the mis program? He is setting himself up to fail by pursuing something he’s not even interested in.

One of the main things experience has taught me is that the world if full of people who talk, but very few that do. Obviously the guy is more interested in gaming than programming. I don’t know but I’ll wager that for most of the people you’ll meet around here the fun is making/trying to make the games, more so than playing them! :wink:

#11

I think he means that if you’re able to succesfully make a Java game, there’s no point in switching to C++. By successful, I mean not a silly pong clone but a complex high-end game.

Great article blahblahblahh, it’s got a bookmark in my required readings folder. And isn’t Sirlin great? A link to his page should be a sticky somewhere on this forum.

He really isn’t closed minded- he just doesn’ t have the exposure to know any different. You don’t yell at a 2 year old who spills his milk by accident, he just doesn’t understand. Perhaps yelling is what he needs though. At any rate, I thank you guys for your help. I’ll keep you posted about what happens. Maybe we can try that note-taking case study.

#12

C++ is great. It’s the first choice if you are about to write something like Doom3 and want to release it “pretty soon now”.

But… something like Doom3 is one hell of intricacy. And of corse you’ll also need a whole team of artists, co-programmers, sfx guys, tons of money and several years of time.

It’s highly unlikely that you could ever do something like that in your whole life. Really.

The truth is, alone or as a small team you can only produce games (within a reasonable timespan) wich are about as complex as SNES or PSX games. You just won’t need 99.99% of the cpu cycles and cutting the edge hardware in order to get a mere 30fps. But that’s ok. It allows you to take lot’s of shortcuts. Highlevel stuff. Highlevel APIs. RAD.

Well, maybe your game won’t run on a p200 (that would be my router heh). Who cares? That “4 year old hardware”, you usually have to take into account are pretty soonish 1ghz machines with 256mb ram :slight_smile:

Btw by the time you are good enough to do something like D3 (if ever) you could also take Java.

If there is any shortcut - take it. If there’s a lib, wich does something you need - use it. Just finish it. If you aren’t satisfied with the performance, profile it, find the bottlenecks and do something about it.

That’s the things you really have to do nowadays if you want to finish something (better then pong or tetris). Since we have more then 16 colors nowadays you can put some of the gained time into getting artistic skills.

Oh and if the game is fillrate limited you won’t get any frames more if you use C/C++. For example the game yuckfu (from a friend) is fillrate limited. I could easily write a Java version of the same game wich achieves a higher framerate by using dirty rectangles :slight_smile:

The contest idea from blah³ is pretty nice imo :slight_smile:

#13

If the point of bbb’s contest idea is to show that developing in Java is faster, then there’s not much of a point - everybody knows that.

But what about better and more memory-efficient?

#14

Also, Funkapotamus, if this “friend” is just a friend (and not your programming assignment buddy), why do you care about what language he likes?
Just wondering why this thread exists…

#15

[quote]Also, Funkapotamus, if this “friend” is just a friend (and not your programming assignment buddy), why do you care about what language he likes?
Just wondering why this thread exists…
[/quote]
It’s as simple as this:

-If he teaches himself C++, he willl fail.
-If I can get him to drop his Java predjudaces, he will move forward towards his goal of becoming a game programmer.

Making a game with Java or C++ is much like making a new car. With C++, you start from the bottom, grabbing raw iron and plastic for the car’s parts. Only after you’ve created the parts can you begin designing how the car looks. With Java, the car’s parts are already made, so all time is spent on car development.

In the end, it doesn’t matter how that car is made just so long as it runs. In the car industry, every company has the ability to make cars that run. However, it’s the cars with the best design and function that get sold.

If I start making cars with Java, I’ll become a better developer because I spent all my time on development. The perfect car developer is the one who develops cars, but knows how to make their parts. The perfect car builder creates all the parts, but knows how to put them together.

He wants to become a game developer, not a game programmer (I might have improperly worded things earlier). If he starts out in C++, he will spend more time programming than developing. Like me, he’ll want to figure out all the neat things he can do. He’ll read Direct X documentation, DOS documentation, stuff about pointers, articles about how to understand the Direct X documentation he just read (^^;)- but in the end, all he’d have done was spin his wheels. Sure he’ll be a smater programmer, but he’ll have nothing developed. If he can resist the urge to tinker around with things, he can just bite the bullet and start to truly develop games rather than program them.

If I can get him to realize this, then I’m sure he’ll be reborn. He needs to fuel his passion or he’ll forever be dissapointing himself (as I hear constantly from him- “I need to get working on stuff…”). This is a deep subject who’s surface only got scratched here. The point was for me to learn how I could better help him as a friend.

He’s not a programming assignment buddy. We’re not at the same level of programming classes, he’s behind.
This thread exists because I care about my friend and his future wellbeing. It’s general discussion forum, I figure this post is better than just shooting the breeze.

I understand that it might be surprizing to see so much talk about the trivial debate between rival programming languages. However, as you might have gathered, my friend has a problem. He’s lazy and irresponsible, but with a huge passion for videogames. He sits on his ass all day and though he’s happy, even he admits that in 5 years, things are going to be bad. This isn’t a discussion about how much C++ sucks, this is a plea of help on how to motivate him on getting his act together. Programming is the way to his heart, and there’s nothing wrong with asking for help on how to make a better argument for why Java should have a home there.

Edit: I accidentally hit submit while I was still writing.

#16

LOL. This is exactly how C++ programmers used to describe C++ as being better than assembler, for all those assembler programmers who believed that C++ was “too slow” (yes, really), and that “all that high-level programming stuff is unnecesary” etc.

#17

[quote] I understand that it might be surprizing to see so much talk about the trivial debate between rival programming languages.
[/quote]
Welcome to the internet, its called a “language flamewar” :wink: More bandwidth has been wasted on the language x vs language y debates than spam ;D

If he wants to be a gamedeveloper he should take some 3d animation classes, there are plenty of programmers around! Could you imagine the reaction someone would get if they showed up on a any gamedev forum with a complete set of models and levels looking for volunteers to program an engine? I should switch disciplines…

#18

[quote][…]I should switch disciplines…
[/quote]
me on page 1:

Low poly (=game) modelling is really tough. So is texturing and general 2d stuff. But it’s nice to have some skills in those areas :slight_smile:

There is also alot of math you might want to learn (I have lots of fun with physic related stuff right now).

#19

You make the assumption he wants to work on volunteer games.

In the commercial world, the standard reaction to a modeller is closer to “so what?”. There are thousands of fine arts and art-school grads each year, and a large chunk of them have made heavy use of CAD (and, to be honest, it doesn’t take that much for the rest to learn enough to become productive modellers).

Sure, if you’re a good artist, and if you go off and practice a lot of animation with the current tools, and learn to program sufficiently well to be able to make complex animations, then you have a lot of value. But generally the artists get less say in the game development than the developers.

Level-designers are a different beast, being more closely aligned with game-designers, and for that your employer typically wants examples of your ability to make “fun” levels more than the ability to model in intricate detail.

#20

IMHO and IME it’s not particularly tough at all, it’s simply time-consuming. If you know how to draw, and know how to model, and can think in 3 dimensions, and have a halfway decent modelling tool (attempting to model without these is like attempting to program without having learnt algebra).