Hacker News Reader

I completely agree. My game dev past in Massively Multiplayer Online Games directly led to work at FANG in cloud engineering. Work on optimizations, storage, comms speed, and stateless flows led to large improvements in product performance. All of that emerged from running into roadblocks on games engines. Even rendering was useful in understanding how to get better performance and frame-rate on client applications -- engineers don't typically think in terms of frames-per-second when writing UI code. My experience in cache optimization for game objects directly applied when used to speed up trading platform transaction speeds. Experience with threading was also given and not common in normal engineering (at least at the time).

I actually think it's the opposite. Nearly every aspect of game development is challenging and transferrable. I can't speak entirely for people who did more of their work in Unity (I was in the 'build it from scratch' generation) but I suspect any high-performance game requires an equivalent level of getting deep down into the code and having a critical understanding of your structures, algorithms, and how your computer primitives fundamentally work and interact.

That alone will usually put you into the upper tier of applicants to any tech company. Your resume, if written correctly, should easily highlight all of this.

> but I suspect any high-performance game requires an equivalent level of getting deep down into the code and having a critical understanding of your structures, algorithms, and how your computer primitives fundamentally work and interact.

I don't think most indie devs developing games are making "high-performance games". Rather they're spending most of their time in the "business logic" of the game, since most optimizations will have been handled by the engine (I don't think this is necessarily a bad thing either!). I definitely feel that I learned more about data structures, algorithms, and performance optimization outside of the game dev world than in it.

And I too completely agree. I agree that I had a head start with learning graphics programming in C in 1996 and have since contributed to developing game engines, games using those engines and most recently (3-4 years back) prototyping the integration of ML models into gameplay. Try doing that with acceptable frame rates and it's a tiring but fun challenge if you like such stuff.

I have swapped between writing games/tools/libraries and working at FAANG the past 7 years and I can say that I get to work as Research Engineer at FAANG's only because of the prior game programming experience. Like anything else, it's not for everybody but game development is very interesting and IMHO cheap way that is at the intersection of some math/physics/performance/complex logic all in 1. This graphics deep dive also gave me opportunities to work in 3D scientific visualisation, augmented reality etc.

The only problem with game development is learning it from scratch/ground up while worthwhile is a long, arduous route; early starters have an advantage and now with game engines like Unity you may not have to poke under the hood but without that poking you cannot master the graphics tech stack. But then we can say that about mastering any subject!!!

> Nearly every aspect of game development is challenging and transferrable

I think thats if you dive into engine development alongside game development.

What I have found is a lot of newer game developers don't know the fundamentals because unity just takes care of it. Instead of spending time learning about data structures and the complexity of building a large interconnected codebase they memorise the unity api and learn how to make 'free floating' scripts that don't have to solve the hard problems.

I used to prioritise game developers when hiring for my corporate job. But I am finding this is less of marker of a good developer. Game development is still hard because it requires a huge amount of knowledge across many disciplines. Games programming has become significantly easier since the rise of unity and unreal.

All you need todo is memoize l33tcode to get into FAANG. b4 that, graduate from a respected uni and u would have a recruiter at your email door setting up a phone interview looking to shape you into a good little employee.

Game programming is hard. its depressing to read because its true for most. This article resonated with my 13yr old self (when i was making q-basic games). Working in the enterprise is cushy if u want heaps of money just do it in Defence and get access to all the awesome tech 10 years b4 Oculus.

Your job as a game programmer readied u for a career at large corperations. Great! Thats not the norm! The norm is a toxic boys club of chaddery.

https://www.nbcnews.com/tech/tech-news/women-pry-open-door-v...

That said, its pretty stupid to think your skills are not tranferable. But also there are new breeds of programmers these days who companies are happy to employ who really offer nothing more then API knowledge.

I think the author just needs a break and a little perspective. As you say they're wildly off the mark regarding transferrable knowledge. They even say

> Especially for an indie dev– you are more than just a code monkey. You will have to know a bit of everything. And even in traditional gaming, most developers seem to know more than their narrow field of coding (just my impression from interviews etc).

at the end. They're clearly aware of the skills they've picked up.

Is the disconnect on skills around the semantics of "transferrable"? Your reply and parent point out that you see lots of skill transfer. But that's "you." What about from a prospective employer's point of view?

Here's a totally different example of the same. I spent 18 years getting really familiar with Smalltalk (specifically VisualWorks). Really familiar. I wasn't afraid to implement goto by mucking with context objects. If I could imagine it, I could do it. Learning and being proficient in Smalltalk has given me huge transferrable skills in modeling, messaging, factoring, abstraction, and so many other things. But telling a prospective employer that one of my skills is to rock things with Smalltalk, isn't very transferable at all. Getting them to accept that I've transferred those skills to be a better Python, Kotlin, C, Linux, Elixir, Swift guy is a long shot. So it's not directly transferrable.

Half the languages you named didn't exist when you started doing Smalltalk. So how can you be handicapped?

I'm not claiming I'm handicapped. What I'm saying is that a "transferrable skill" is a vague term. We can, if we put our mind to it, transfer skills from just about any domain to another. But just because we can put it (the skill) to continued use, in an altered domain, doesn't meant that others will value it as a transferrable skill. It's kind of like trying to transfer credits between colleges. If they don't recognize the skills you've acquired at your former school as transferrable, you don't get to transfer them, and have to retake classes that may seem pretty redundant.

So, transferable yes, but marketable no. I would classify those as different things.

Because the employer judgement is often shit. So becoming an absolute baddass in $obscure_language isn't good for your CV, even if it can be great for your skills as a programmer.

> Getting them to accept that I've transferred those skills to be a better Python, Kotlin, C, Linux, Elixir, Swift guy is a long shot

It may just be me, but I wouldn't doubt that your Smalltalk depth has made you a better developer in general, and that applies to any language.

Agree. I was in the games biz for 15 years and found that the skills I learned were highly sought after by non-biz companies. Especially knowing how to make software run fast and being able to invent and implement unique solutions to hard problems that you can’t just download a framework or library to fix. I will also say that the smartest and best developers I have ever worked with was in the games industry. However the pay and life balance is better outside of the games industry so I switched.

>being able to invent and implement unique solutions to hard problems that you can’t just download a framework or library to fix.

The first step on this is really learning my algorithms and data structures, no?

I'm a 'needs frameworks' programmer at the moment and I often hate what I do because programming by scouring reddit for something that isn't well documented is beyond demoralizing. If I could just build what I need on the fly that would be luxurious.

Yes you are right. That’s what you learn at good universities: all the core data structures and algorithms you can use to build whatever you want. However you don’t need to go to University to learn this. There are great books on algorithms and data structures you can read to learn it as well. Playing around with different algorithms/data structures in your own non-work code is a fun way to learn.

Another fun way to learn this is to make your own rogue like game. There is a great online community of people who help each other building rogue like games. You learn a lot of interesting ways to think about algorithms/data structures by doing that. And it is fun!

Good points.

I knew a Flash dev who failed miserably after Flash went down the drain.

I also knew a bunch who switched to other tech no problem.

I know a bunch of smart people who never finish anything, but I also know many that do.

As a former semi-prolific Flash game dev I will say that nothing has really compared to working in that and I can no longer crank out games like I could when Flash was the big thing.

Although that being said, part of that may be my standards for what constitutes a finished game have gone up a lot since then, probably.

Hey thanks! I've been excited lately by building up a skillset for (hopefully) an entry-level graphics programming job because I figure it'll have a high intellectual ceiling. I almost read this article.

I'm just getting into game dev as a hobby and stopped reading it for the reasons you stated :)

There are problems that exist though. As a general creative stepping into designing vr environments in unity(which is just a Game) I want a platform to attract users from, I want a demo gallery where a user can step into my world and try it out for a few seconds, like walking around a mall. I think we are still a bit away from that sort of exposure/tooling/marketing/platforming for indie devs.Its just sooooooo much work to create a project like this and get people to experience it. I guess I could be making a Roblox game... But that just doesn't have the freedom or portability.

Seconded. I also found my skills transferred easily outside of games themselves.

To be fair to the author,

> More like "The Harsh Truth About Me"

This is true of *all* advice

A very difficult aspect of indie game development is that the quality of the end product is easy to discern. We all can tell if someone creates a game that's got good graphics, sound, gameplay, is fun, etc. The indie developer who does all that has a great deal of skill in multiple areas. The harsh truth is that not everyone can be that good.

My daughter when she was around 6 years old loved to put on princess dresses and dance around singing. In her mind she looked and sounded like the cartoon characters in the movies.

One day, I pulled out an old VHS camera and set it up on a tripod for her. She danced in front of it for half an hour, recording a bunch of songs. Then she sat down to watch it. After the first few minutes, she turned to me with tears running from her eyes and yelled, "I stink!" She was inexperienced at dancing and singing, but she knew immediately from watching herself that she wasn't very skilled.

There's also a subjective aspect to art.

A game that I find tedious could be someone else's favorite game.

Your daughter realized she didn't meet her own standards of good, but that doesn't mean she would agree with everyone else on what singers are good or bad.

There's a great audio clip from Ira Glass about the gap between your taste and your ability.

https://vimeo.com/85040589