The new forums will be named Coin Return (based on the most recent vote)! You can check on the status and timeline of the transition to the new forums here.
The Guiding Principles and New Rules document is now in effect.
Breaking into programming - postgraduate
Nissl
Registered User regular
Registered User regular
So after four years I'm realizing I don't like my job now and don't think I'll like where it will take me in the future. Time to start looking at alternatives while I finish off my Ph.D. over the next 18 months or so. (I won't come out in debt since my position pays passably, so I think it's best to show I can finish what I start, especially since I'm so close to graduating).
One class I really did enjoy and do well at in college is programming, I took a "programming for beginners" class where I learned to work with Python and built up to making a tic-tac-toe progam (actually made it unbeatable, just for fun.) So I want to get back into programming, and if I find I still ike it I want to build up to getting an industry job, ideally in the gaming industry. I can see myself eventually climbing the ladder to a position making/implementing core design decisions, particularly in strategy and RPG games; I'd love to see if I could build a better mousetrap. Yeah, I know that's a position that will take me a while to get.
Anyway, what languages, what approach, what books to start with?
The suggestion I got from my friend in the industry is start with Java (a little higher level than C#), Eclipse for software (free and good?), and a book like "Learn Java in XX days" (I really want a book that gives you homework, I learn best by doing). Then build up a portfolio, make a small 2d game and/or some mods to show off. I'm a little worried about how to approach that because I have shitty art skills and hate drawing, but it also sounds like interviewers want you to do all your own work. I could also always take extension classes if those are needed to show expertise, but my intro class was ungodly slow and the classes at the best local school (U Cal-San Diego) are $650 a pop.
Thanks in advance for any advice!
One class I really did enjoy and do well at in college is programming, I took a "programming for beginners" class where I learned to work with Python and built up to making a tic-tac-toe progam (actually made it unbeatable, just for fun.) So I want to get back into programming, and if I find I still ike it I want to build up to getting an industry job, ideally in the gaming industry. I can see myself eventually climbing the ladder to a position making/implementing core design decisions, particularly in strategy and RPG games; I'd love to see if I could build a better mousetrap. Yeah, I know that's a position that will take me a while to get.
Anyway, what languages, what approach, what books to start with?
The suggestion I got from my friend in the industry is start with Java (a little higher level than C#), Eclipse for software (free and good?), and a book like "Learn Java in XX days" (I really want a book that gives you homework, I learn best by doing). Then build up a portfolio, make a small 2d game and/or some mods to show off. I'm a little worried about how to approach that because I have shitty art skills and hate drawing, but it also sounds like interviewers want you to do all your own work. I could also always take extension classes if those are needed to show expertise, but my intro class was ungodly slow and the classes at the best local school (U Cal-San Diego) are $650 a pop.
Thanks in advance for any advice!
360: Purkinje
Nissl on
0
Posts
Anyway if, like me, you were directed to the Python website during the course of your studies, you may have seen this starting-from-scratch tutorial in Python that also shows how things are done in both Java and VBscript. I know nothing about either of those (and I've only just started this Python stuff) so I don't know how helpful it'll be. From the looks of it Java and Python are pretty similar.
http://http://www.freenetpages.co.uk/hp/alan.gauld/
To answer your question: You should learn C. Video games are written in C++, but many are written in a way that it just looks like plain C code. You need to know C because you have to understand pointers and lower level issues that Java/Python and others will abstract for you. Video games need to run as quickly as possible, so that sort of abstraction isn't allowed. You will find C hard, time-consuming and archaic, but it's best to just jump right in and get to it IMHO.
BUT
I really do not think you have thought this through at all, I think being so prepared to throw away 4-6 hard years of work is crazy, and people who interview you will too. Is there no chance you could use your PhD to do video game research as a postdoc in academia? You can steadily build up your programming knowledge, and hopefully use some of the things you have learnt in your current studies, as most video game research is very cross-discipline. But I guess it depends what your subject is. Then you might be able to get hired in video games at a higher level and get more money in the long run, or work towards a faculty position.
I agree with this mostly, except you should probably just learn C++, and not really bother with C, because, well, theres not a huge point to learning just C.
However, if you're looking into making games, realize that it takes a LOT of time and skill to learn not only C++, but how to use graphics libraries to create real-time games. The thing is, programming as a hobby can be fun for you, but you'd hate programming for your livelihood.
C++ is pretty much C except with classes added on. Most of what you would learn about C applies directly to C++ too, so there's no point in skipping it. You're probably better off being able to immediately handle C code if you come across it.
As a bit of a warning, though, learning C first might make it difficult to train yourself to take full advantage of object orientation in C++. However, you can probably cure yourself of this by learning Java (or maybe C#, I don't know that language at all), which shoves object orientation down your throat and beats you until you swallow it.
As for books, I've never liked anything that has a title of learn in x days, because I find they're alright for going through the first time, but become useless as a reference. I prefer something reasonably thick that covers more detail, describes all the basics up front, then goes from there. Ivor Horton's Beginning Java 2, JDK 5 Edition did a pretty good job of this for Java, so long as you can live with a few errors in the sample code. I don't know any great C++ books off the top of my head, but I haven't looked in a really long time.
If you ever need to talk to someone, feel free to message me. Yes, that includes you.
Oh, and plan on spending the next 5-7 years learning this stuff at the hobbyist level before you're even remotely qualified to apply for a junior programming position at a game company.
The PhD is in neuroscience, and I really don't like the academic environment or most of the job functions that being a professor entails (and I've set up and run my own project, my PI has been quite generous in that regard, so I've seen pretty much every aspect of the job firsthand.) I feel like it's a poor use of my talents and I want to do something that involves more... building. There's no way I'm going into patent law, so the last clear use of a PhD is climbing the ladder at a biotech, but the hands-on won't be any more fun on the way up and I don't think I'd find the business end particularly fulfilling either. Video game research is an interesting idea, and one I will think about heavily, but again I want out of the academic culture yesterday and I suspect I'll be in the same place after 5 years of ladder climbing as after 5 years of research.
I'm not worried about the lifestyle; I'm working 50-60 hours a week already, and entry-level coding jobs pay competitively with post-PhD pre-professor academic jobs (maybe it shouldn't be this way, but my friend's entry-level world design job pays $10k more than the next rung up my ladder). Also, my family situation is such that I'm well off enough to do anything, just not well off enough to do nothing, so I don't care if I'm capped a little lower than a university professor (it's probably the reverse, though). And not to turn this into a dicksizing thread, but I graduated highest honors from Harvard in a major full of premeds, and have a perfect GPA at a top 5 grad school in my discipline. I'm not worried about coming in a little late and being good enough to compete and get promoted.
But I do understand the portfolio is more important than that stuff, which is why I want to build up a portfolio, make some mods and/or build a small game or two in my free time while finishing grad school. I'd most likely be eyeing a 2d action/adventure game or a mod for the new version of UT when it releases this fall. If I don't like it, then I can take the PhD and go the industry route. My question isn't so much "this is my plan" as "this *could* be my plan if I take to programming in my free time over the next couple of months, where do I jump in that would let me build a portfolio quickly and put the most marketable languages on my resume later?" If, as zilo says, I should resign myself to slowly building up a bigger portfolio, then I have fallback plans. I also have multiple personal industry connections, indeed I've been told I could walk in off the street and get a QA job at a chinese MMO firm right now if I wanted, so I'm hoping I'll be able to take a little bit of a shortcut at some point.
Be realistic. Get a job and work on this stuff in your downtime. There is literally no chance that you can learn everything you need to know, starting from scratch with C++, in less than a few years.
In fact, small software companies are looking for guys exactly like you: Very sharp that can learn easily that can pay you somewhat decently. It's a bit difficult because you're thrown into the mix with people who have very different levels of experience but it's a great way to learn. Read and code in your spare time to catch up. Once you get the experience, search for larger companies or start contracting on your own!
Just a thought. Game programming is a large commitment and difficult to break into.
Thanks for the advice, this is really good information. Do you think it would be a worth it to take extension classes in my free time, (I could probably do the intro class, finish 4 quarters of C++ and maybe 1-2 of another language before I graduate) then go from that into a master's program? I can get the financial backing to do a master's if that's the way to go.
Another option I will definitely keep in mind, thanks.
Also, thanks for the idea jothki. Right now I'm not using my computer for anything other than basic stats and writing, but I will keep an eye out as one of my future projects has a programming element.
Also, does anyone have a good starting point to learn about scripting levels/behavior? Can anyone comment on the requirements, upsides/downsides of this position?
Of course I'll do lots more research into this on my own, but these forums are a great resource for spitballing ideas. Thanks again.
As for scripting, you already know Python so learning other scripting languages shouldn't be too big a stretch. For a well-featured and cheap playground to work with, give the Unreal engine a shot. If you can find the UT2004 Editor's Choice edition, go for it. It's got a DVD with a bunch of tutorial videos that're pretty good. There's some good tutorial type videos on 3dbuzz.com too.
In a dev studio scripting jobs usually come as one of two flavors: designer (aka map-maker, world builder, content maker) and programmers with free time (or super-junior programmers). Design jobs are imo the hardest to get because everyone thinks they'd make a great designer and every entry-level job gets thousands of applicants.
Remember to start small. One of the biggest reasons people burn out or lose their desire to keep at it is by biting off too much too early. Pick small projects that you can complete in a weekend initially, and once you get your feet under you give some thought toward making a portfolio-quality project (that you'll probably trash and rewrite three times as you go along). If it takes less than 4 or 5 months to complete your portfolio, you're probably not doing it right ;-)
edit: oh yeah, C# and Java. They're neat languages, but they're interpreted (read: slow). As it matters for game development they're really only useful in tools development. There are some niche markets that use Java as a primary language (mobile phones, web games) but by and large C++ is the mainstay of the industry and will be for a long, long time.
Don't get me wrong, the C#/XNA stuff is grrrrrreat fun to play with. The Xbox portability is a fantastic feature, it's just not on track to replace C++ in any major development house. It's intended for hobbyist or indie game programmers.
Page after page of incredible. Plus the author just bleeds credibility.
3DS: 1650-8480-6786
Switch: SW-0653-8208-4705
My only complaint about the book is that the title makes it blend in with all the crappy "Learn Game Programming in 9 Days!" type stuff. It does nothing to let potential buyers know how full of win and awesome it is. I stumbled across it in a Barnes and Noble a few weeks ago, I felt like I'd found a 10-carat diamond in a big pile of turds.
Promoted to where? People in the games industry don't make very much. You might make less as a postdoc, but as soon as you get a faculty position and head for tenure, your salary begins to rocket. The years of crap pay balance. That is not true of the games industry, games designers are not rolling around in money, unless they managed to sell a start-up, in which they had a large amount of shares, to a publisher. I would hazard to guess there are less than 50 or so of those people in the world. I'm concerned you seem to have a warped view as to how well people are doing in games. I have super-respect for anyone and everyone who has the talent to work in games, but they have chosen to make less money than business IT for the love of games. For you, you'd be dropping down not one, but two rungs of the pay scale, as your PhD should open doors for you at higher levels. Particularly in BioTech, which could not be more financially solvent right now.
Top 5 Grad School Material in a PhD as tough as neuroscience... it just seems such a waste, and you are never going to be paid what you are worth in the games industry. You could be researching AI neural networks in video games or something, you'd avoid graphics programming, you'd use your current PhD skills, you'd get to work with video games.
But anyway, I'm going off-topic.
I've got a buddy wrapping up his doctorate in computer science. I don't bother commiserating with him about low salaries in our chosen field.
edit: I've got another friend working in the industry as a programmer who makes overtime. Overtime! Amazing!
we also talk about other random shit and clown upon each other
Man, Michigan's economy must REALLY be skewing things. I switched from writing code for WebMD in Ohio for peanuts to working a helpdesk for a modestly decent wage, of course this was post 9/11 and dotcom bust. Most of the coders I knew in Michigan were starting out around $35-40.
I've managed to get some good promotions and raises since then but it would be nice to use my degree again. Time to break out the compiler and shake the rust off my NURBS.
If you can get the scripting (Perl, Python, etc) and graphic libraries down, there's also the potential to work for a CGI company creating utilities for 3D programs like Maya.
Hrm. I was making $45k a year doing web programming stuff in Nebraska back in '02, $35k for a C++ programmer is painfully low.
Although I did get offered essentially the same job in Dallas for $35k, which I turned down.
I would go so far as to say the 3d math stuff is necessary for game programming. You have to do a huge amount of work playing in different coordinate systems and the like; learning the derivations of transformation matrices and stuff really helps keep all that straight in your head.
I second OpenGL, too. The OpenGL Programmer's Guide (aka the Red Book) is a great resource, although it's more of an intermediate to advanced text. Ditto for the OpenGL Shader Programming Guide (aka the Orange Book).
So, uh, I dropped out of a PhD program in neurobiology with a master's and worked in games for a couple years. I was lucky, got some gigs doing writing and design. Had fun. Now I'm doing patent law. Which, I know, you said you'd never do. But, really, the hours in law aren't any longer than they are in games and the job is just about as interesting. Games are shiny for a while, but it's a job and there's a reason the industry burns a lot of people out by the time they're in their early 30's. And, uh, the base pay in patent law is on the order of 2-4x what you'll get in games (your first year out of law school, the differential grows rapidly after that, in favor of law). Plus, you get a 3-year vacation in law school, which will require much, much less effort than what you've been doing. Seriously, go to UCLA (or similar) and spend three years surfing (or similar), graduate, and profit.
Just stuff to think about.
Something that officially recognises when you bring the knowledge on threads.
we also talk about other random shit and clown upon each other
B.net: Kusanku
Another good story I will keep in mind, though I will almost certainly go the MBA route if I decide to do this. Watching my friends in action, an MBA is a cakewalk to get even at the top schools, with my background it will pay very well, and I find economics and business decisions interesting and law insanely boring. (I've had plenty of exposure to both as my father is a law professor and I was an econ major for a while.)
Thanks for the large doses of cold hard truth. I started this thread since the search engine results seemed to be mostly from undergraduates, and therefore about majoring in CS/graphic design, taking classes, and what to do after graduating school. I certainly don't think I suddenly realized my calling and now everything will work flawlessly; I just want to explore efficiently for the next year and a half as it's becoming more and more clear that academia is not my bag. I guess I also don't want to wind up locked into a family and career at 40 with any remaining "what if"s, but that's another story.
Anyway, I'm going to pick up those recommended books, sign up for an extension class and see where things take me. If I decide it's not for me I can take the PhD and go into biotech, add an MBA in a couple of years and make cash money money.
I'm actually incorporating some programming into my Ph.D. work now. I'm doing some pretty basic stuff in Delphi regarding alterations in neuron spacing in the brain (with manual identification of cell locations), with the help of our lab's resident programmer. I'm taking a trip to China that would knock out the first 2+ weeks of class this quarter, and I have several huge projects "due" right after I get back, so I'm going to delay taking an intro C++ class to next semester.
We'll see where this all leads; I'll try to toss in more updates every month or two.