Staffs Uni visits Develop 2014 – Day 2 Develop (Part 2 of 2)

Following lunch at our second day at the Develop Conference in Brighton we attend some vary different talks, from highly technical to very business oriented.

Rendering Fields of Grass Using DirectX 11 in GRiD: Autosport

Presented by Richard Kettlewell
Attended by Paul Boocock

This talk was far more technical then many we attended, with Richard discussing how Codemasters implements the fields of grass that can be seen in GRiD: Autosport.

16a0b255b6acdf64320268d11b6aaf04_8782Whilst, I won’t be going into the finer details of his talk (as it was far too technical for a couple of paragraphs mid blog post!) there were some key points to take away. Traditionally grass rendering has been done with billboarding as it’s a pretty expensive operation. However, with the advent of DirectX 11 and Compute Shaders there are new options available to render grass.

Codemasters took inspiration from implementations shown by Outerra (http://outerra.blogspot.co.uk/2012/05/procedural-grass-rendering.html) and nVidia (http://blogs.nvidia.com/blog/2010/03/16/new-directx-11-technology-demos-show-future-of-game-graphics/) and they found themselves with the following goals:

  • Keep all data on GPU
  • Soft Grassy Texture – Get rid of polygonal look of terrain
  • High Resolution Textures
  • React to dynamic scene objects (Wind, cars driving over, billboards falling over)

Ultimately, they used a mixture of different techniques that made use of custom shaders in DirectX11, level of detail techniques yet still retained bill boarding as they had tight performance constraints. They also managed to implemented shadows and managed to get real time deformation using the skid marks engine to render the skid marks into a texture that adjusts the height map and then changes the grass billboard positioning.

There’s far more technical discussion around this and potentially we might hear more about this in the fast approaching academic calendar (no promises!).

GAMESbrief and Bartle: A Conversation About the Future of Free-to-Play

Presented by Nicholas Lovell and Dr. Richard Bartle
Attended by Paul Roberts

This was an interesting discussion around F2P and the future of it. There are two very differing opinions on F2P with many loving it and many being strongly opposed to it. This was an entertaining talk that I’m sure left many in the audience considering both sides of the F2P argument.

There was a lot of good discussion around whether F2P works and what sort of systems work best in this type of game. It’s interesting to consider how F2P games can be made without causing issues with immersion, often users are asked to buy or purchase something and that feels unnatural in the current game play. Developers should be trying to make use of F2P as a technique to enhance their games rather than just using it to monetise as it doesn’t always work (and there’s many examples of bad F2P games).

Indie Boot Camp: 10 Ways to Guarantee NoOne Will Ever Fund Your Game

Presented by Caspar Gray
Attended by Paul Boocock

Caspar went through 10 things to make sure you do if you find yourself in a meeting where someone wants to fund your game but you really don’t want that funding… (Very amusing!)

#1 Over confidence

Biggest problem is ambition (First game is an MMO!)
No need to plan for the long term!

#2 Giving up / No Effort

If it’s difficult it’s not worth doing!
Why bother with a website!

#3 Failure to identify audience

Your games for everyone!
Don’t pay any attention to your competitors

#4 Content Choices

I could make a better version!
Niche Subject Matter
Add nothing to the genre
Unappealing characters, offensive stereotypes
Everyone loves controversy
Naming!

#5 Failure to Schedule

Lie! Just randomly set a release date and development schedules
Polish? That’s for furniture! We get stuff right first time!
What Testing?

#6 Jack of all trades

You can do it all!
No need for specialists
Use Open Source that shouldn’t be used for commercial

#7 Poor Materials, Poor demo

Include lots of story and don’t talk about the gameplay
Don’t highlight key features or innovations
Don’t update your documentation or don’t make any documentation
It’s Alpha!
“It runs fine on my PC”

#8 Presentation Skills

Don’t tailor to the audience
Don’t practice and overrun
Never take a presentations skills course
Don’t ask questions

#9 Trash Talking

Slag off other developers and games – Especially successful ones!
Your opinion is all that matters

#10 Poor negotiation

It’s never too late to derail!
Say different things in meetings
Make lawyer jokes! (Especially when a lawyer is present) 

I Will Survive

Presented by Jo Twist, Alive Taylor, Anna Marsh, Ella Romanos and Fiona Sperry
Attended by Paul Roberts

We don’t have many notes on this talk but it was predominate abou working as a small business and how to gaining funding.

‘Surviving and thriving as a small, ambitious UK games business is tough. With studios trying to balance work for hire and their own IP generation, what are the strategies for surviving that are working for small businesses? Has the investment culture improved over the last year? What support from public policy do we need? How do we increase resilience and change business practices to flex with a time of transition which looks like it is here to stay.’

You can find lots of information relating to this talk at: http://ukie.org.uk/

Playing the Security Game: Lessons Learned From the Game Security Trenches

Presented by Wade Winright
Attended by Paul Boocock

Wade works as head of security at Lift London, a new Microsoft Game studio working on mobile games and he discussed many interesting things they do on a day to day basis to ensure their games and especially their backend systems are as secure as possible.

One interesting element of what Lift are doing at the moment is they are developing Games as a Service using cross platform tools (Unity) and Azure as their backend. This allows them to write less code, as they get great C# code reuse between the client and the server. Four core elements Wade briefly discussed during his talk that Lift use during their development:

  • Data Driven Development
  • Hassle free continuous deployment
  • Analytics lead features and improvements
  • Embracing the cloud

He also discussed whether data can be trusted, more so the data that is being sent to the server from the game. The answer was pretty much a no. There are various tactics that can be employed to try and verify the data or protect it however many add complexity or to the amount of data that needs sending (like obfuscation). Basically don’t send out data that you don’t want people to modify was the motto of the story. Lift are also doing some clever server side verification, to ensure the commands being sent from the game could actually happen in the game world in the timeframe and order they have been sent – refusing to accept them if the server deems the combination of events to be impossible.

There was also talk of a treadmill concept to security – where you build multiple layers of protection and when one is defeated you build another and then you continue with this cycle, like a treadmill. Systems will ultimately also be broken or cracked so you just have to stay one or four steps ahead.

Leave a Reply

Your email address will not be published. Required fields are marked *