GDC 2013 Mon Amour

I have a love-love relationship with the Game Developers Conference, since 2006 when I had the chance to give a talk and experience first hand the joy of talking in front of a flock of super techie people staring at me and waiting for my first slip up (and they happened). At the end of that talk I was rewarded by a guy from NVIDIA with a compliment I still cherish: "Fran, I loved how you talk, the rhythm, your pauses, that was good man". I blushed and I was too embarrassed to reply that my manner of speaking was nothing else than heavy jet-lag kicking in and making me literally fall asleep on stage every few minutes. A job well done.

This GDC was ok, mainly because I met some old good friends, from Georg Backer to +Ury Zhilinsky, to the guys at Crytek Kiev and Frankfurt: I can't say I miss Frankfurt, but it was good to see them again.

The talks I attended were ok, nothing spectacular. Clearly mobile gaming is a big big topic, but, from the purely technical point of view, being told in 2013 how blooming is implemented feels a bit old: basically mobile graphics is where desktop graphics was around 2006, it needs the big guns to step in and show these kids how to do shit. The mobile GPUs, though, are getting pretty darn fast: ARM reported that mobile GPUs are almost on par when it comes to computation power with console GPUs (again, about 5 years ago tech), but lose big time when it comes to available bandwidth: accessing external memory takes a lot of power that is a scarce resource in the mobile space. Last take away from this talk: in mobile, deferred GPU kick the proverbial ass.

That leads us to the NVIDIA talk and Tegra 4: Tegra 4 looks very powerful on paper, but it also looks like it could easily be used to warm up your cold winter nights in Europe. An immediate rendering architecture with non-unified shaders is not what you can call power efficient. But it's fast.

But enough with this technical mumbo-jumbo and let me introduce the absolute protagonist of GDC 2013, the undisputed star of the moment, the return of the king... suspense... PROJECT SHIELD.

Applause please

I thought that Microsoft Surface was a pointless device, but Project Shield wins easy-peasy the first price for pointlessness: it doesn't make any sense whatsoever, it's gimmicky, it's ugly, it's cheaply built, I had the "pleasure" to hold one and it felt as it was made of plastic. In fact, it is made of plastic. And it's butt-ugly. Opening the screen makes a weird sound as the hinges weren't properly oiled.

How they can think anyone will buy this is beyond me.

Final word about John Romero, who I met the last day while walking out of a restroom (please, no jokes): this guy rocks, he's nice, approachable, funny, and bright as the sun. He was the highlight of GDC 2013.

I want to learn game programming #1

So you really want to learn game programming, and you are asking where to start. I've been on the receiving end of this question so many times, that I tend to answer in autopilot. Let's get the scare talk out of the way first: making videogames is neither glamorous nor will it make you (or me) rich, it just so happens to be our passion.

It's hard work, there's a ton and more knowledge to build up on the technical side before even trying to attempt making a very simple game, and, after that, there's a lot more to learn about how to interact within a game team, with colleagues coming from different nationalities and working in different fields (from programmers to artists to producers to... managers). Work hours are long, depending on where you ended up they are even longer, the dreadful promise of crunch time in the hope that the game will be shipped on time is an always looming menace, with the little solace of the faint hope of a bonus at the end of the project (that practically very rarely or better never materializes). Speaking of money, game programmer is a very low paid job compared to other engineering fields.

If you are here to make money, move away. There's lots of good blogs on the web that teach you databases.

Are you still around?

Next question to answer is: what programming language do I learn?

Quick answer: not C++. Pretty much anything else is fine, but if I have to choose, I'd go for either Python or Ruby, two high level, dynamic languages that take most of the gritty details out of the way while you build up your confidence in writing code. At the beginning especially, you need to focus on learning how to solve problems in a simple manner and how to manage complexity: programming is hard, programming is a blend between engineering and art, programming is a creative and artistic endeavor. Mastering the Art of Programming, being games or any other engineering field, requires years of practice and not everyone will likely succeed: Keep It Simple must be your mantra.

Why not C++? Because it's a bloated, difficult, over complicated, extremely powerful language that will keep you bogged down forever in details that are completely irrelevant to learning programming. You will have to master it one day, but that day is not today.

... to be continued

VFX protest

Making visual effects for a movie is surprisingly similar to making a videogame: the workflow maps very closely, both need graphics engineers, modelers, animators, technical artists, producers, more or less the same process. Historically there has also been a fairly consistent exchange of professionals between the two in the last ten years. I've always looked very closely at the VFX industry to gauge what might happen to the Game Industry, for this reason the news of the VFX protest struck me pretty hard: it's about time there's a protest!



To cut a long story short, many of the visual-effects makers of Life of Pi, who made the creative vision of the award winning movie possible, were let go without compensation for them or their families.
These people are not cheap, they bring with themselves years of technical and artistic expertise that is extremely hard to build up, they work hard, long hours, weekends after weekends with the promise of a bonus after the project is done; a bonus that very rarely materialize. More often than not they are trimmed at the end, by a very competitive industry that is not sustainable in the long term.

Rings a bell? Yes, the same happens in the Game Industry.

Like the VFX industry, the Game Industry is very badly organized, with lousy project management demanding crunch time after crunch time even when all evidence accumulated in a hundred year of research on the matter suggest that it is counterproductive. And like in the VFX industry, workers are let go easily at the end of the project.

People working in the VFX industry and in the Game Industry need to draw a line in the sand, organize themselves and demand sustainable business that rewards who give life to the next Life of Pi or the next LA Noir.

How? Give me some feedback.

XNA is dead - Oh noo!

Actually it's been dead for few weeks, but it took me some time to get over my mourning for what has probably been one of Microsoft's best ideas. The API was well thought, clear, nicely designed and, most of all, simple yet pretty versatile.

So they killed it.

I used to teach XNA Studio at the University back in Germany, my students loved it, I loved it because I could visually show interesting game programming concepts in a very slick environment that could target Windows, Xbox 360 and Windows Mobile, with some very good assets management for small games thrown in the mix. it was a great tool to learn game programming.

So they killed it.

I bet they needed resources to focus on Surface.

Playstation 4

And so Sony announced the Playstation 4. The first thought is that the cell processor wasn't that revolutionary after all. The second thought is that indeed the cell processor sucked precisely as foretold years ago. In fact they announced a PC.

The actual show was very much underwhelming, set aside Alex Evans who is always a sexy beast: Alex, I miss you and our late coding evenings and our chats at the hairdresser in Guildford, my man... I don't miss the english rain though.

But the really scary thought is that my Macbook here will outperform this PC, i mean, this console. My iPad will almost as well.

I must not mention Apple on this blog... I must not mention Apple here...

My biggest hope is that Sony doesn't go bankrupt after launching the PS4, cause I have almost 10.000$ invested in Sony photographic gear: I love my A99V and my Zeiss lenses, please Sony, don't die.

Undeniable logic

One of the biggest mistakes I used to make in my career in the Industry was to think that a logic argument, well reasoned and with plenty of data to back it up was enough to convince reasonable and clever people, the kind of people I work with. An example is my favorite topic (no, it’s not commenting code), automated testing and unit testing: we all know that writing unit tests is a Good Thing™. We all do, right? There’s plenty of evidence and my argument for writing unit tests before even writing code is very solid. I was naturally expecting that after some resistance due to various reasons, the vast majority of my co-workers would eventually give up and embrace my reasoning, if not flat out try what I was advocating. But discussing automated testing here is not my goal now: my point is that very few people could actually see and understand my point. How frustrated I had been before realizing that it was all my fault.
It took years to understand that the undeniable logic in an argument is not enough to convince someone of its validity: it’s as important and most often even more important how the argument is expressed, how it’s told and explained.

John is a very clever programmer, extremely sure of himself, who likes to drill into the details of a problem; his method is tried and true, and he wants to give something else a go by himself before incorporating in his work. He likes to picture a problem and see a solution. I approach John and explain him that according to several studies writing Unit Testing  makes code more reliable and reduces cost of ownership for a code base, so he can focus on the general problem rather than fixing small bugs; doesn’t it sound correct? John thinks “bah” and moves on.
Trisha is a very clever programmer, who likes to keep herself informed by reading about how others in the Industry have solved similar problems in order to possibly produce something better for herself. She likes to think about the big picture rather than wasting time on irrelevant details. She wants to feel a problem before touching the right solution. I tell Trisha that if she tries Unit Testing for herself she will find that her code will be even more reliable, simpler and easy to change when she needs to change the details of the implementation, which is a result that she can immediately see after only few weeks. Trisha politely smiles at me thinking “bah” and moves on.
Things are not as simple as in my example, but we can immediately draw some interesting conclusions from it: in both cases what I told to John and Trisha is true and verifiable, but I failed to understand how they want information to be presented to them, which is radically different! John doesn’t care that others find Unit Testing useful, he wants to try for himself, while Trisha doesn’t want to try something if others haven’t already established it’s viable. John wants to see an argument not hear it, Trisha wants to feel it not see it.
Interestingly, addressing a group of people makes it even more difficult because different personalities are likely to be present, how about:

Writing unit tests before code is proved by different studies to be effective, which is something that you can see by yourself in just few weeks of trying it out, either on some small units or in a big problem that sounds difficult to tackle. You can easily feel the advantages or come to ask me for advice if you encounter problems.

How does it sound?
This realization makes leading a group of people even more challenging: now it’s important not only to have a solid and well thought-out vision, but also to understand the nature of the people we work with, how they want information to be presented to them, how they like to reason and respond to problems. Failing to communicate our vision is our mistake as leads.

Post also available on AltDevBlog.

Crunch… crunch… crunch... #2

 

Old and very well known article, but it’s worth a refresh cause it seems to be overlooked:
http://archives.igda.org/articles/erobinson_crunch.php

In most times, places, and industries over the past century, managers who worked their employees this way would have been tagged as incompetent — not just because of the threat they pose to good worker relations, but also because of the risk their mismanagement poses to the company's productivity and assets. A hundred years of industrial research has proven beyond question that exhausted workers create errors that blow schedules, destroy equipment, create cost overruns, erode product quality, and threaten the bottom line. They are a danger to their projects, their managers, their employers, each other, and themselves.

“Video Games Lead to Faster Decisions That Are No Less Accurate”

 

http://www.sciencedaily.com/releases/2010/09/100913121656.htm?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed:+sciencedaily+(ScienceDaily:+Latest+Science+News)

ScienceDaily (Sep. 13, 2010) — Cognitive scientists from the University of Rochester have discovered that playing action video games trains people to make the right decisions faster. The researchers found that video game players develop a heightened sensitivity to what is going on around them, and this benefit doesn't just make them better at playing video games, but improves a wide variety of general skills that can help with everyday activities like multitasking, driving, reading small print, keeping track of friends in a crowd, and navigating around town.

Enuf’ said. Now I can play Starcraft II for as much as I like claiming it’s for my own good.