Slashdot Mirror
Brian Hook Interview
A reader writes:"I just read this very in-depth interview with Brian Hook on a site called Curmudgeon Gamer. Hook used to work for id Software (Quake 2 and 3) and later for Verant (Everquest) and he apparently worked on Glide for the old 3DFX cards. Now he runs his own smaller game company called Pyrogon. In the interview he talks about development styles of Q2 and Q3, MMORPGs, the lessons of 3DFX, and development of cross-platform games like his Candy Cruncher (which is available for MacOSX and Linux!). He even gets into some criticism of modern games and the life of a smaller game developer. Lengthy read, but lots of stuff to think about there since he tells it like he sees it."
Brian Hook didn't "apparently work on Glide" - he was the person responsible for initially implementing the GLIDE API.
Give me a new game genre. && I'd rather do something peaceful to reinforce my feeble humanity.
Here ya go
my religion lies somewhere between buddhism and super monkey ball - pamphlet?
It's time for another interview here on Curmudgeon and this time we have with us Brian Hook, president of Pyrogon Games and former developer at id Software and Verant/Sony Online Entertainment. Before working on the seminal shooters Quake 2 and Quake 3: Arena, Brian was the original architect of the Glide API used by the 3dfx Voodoo line of video cards. After departing id, Brian worked as a Senior Technology Architect at Verant, concentrating on development of technology for next-generation massively-multiplayer online games (MMOGs). After founding Pyrogon in 2000, he created the puzzle game Candy Cruncher, which we reviewed earlier this year. Two additional Pyrogon games have appeared since then, NingPo MahJong and Letter Linker, both available for Windows and MacOS X. (A Linux version of NingPo MahJong will be shipping in the near future.)
This interview covers a wide range of topics: game design criticism, OpenGL/D3D, making money as a smaller game developer, and the importance of porting software to different platforms. Indeed, something for just about everyone. It's quite long, so you might want to bookmark it and consume it in more than one reading.
The interview questions were prepared with the assistance of regular CG authors ruffin and michael.
jvm: What kinds of games do you enjoy playing the most?
jvm: So you play mostly on a PC, as opposed to a console?
jvm: Could we get an example of a game with these egregious design flaws, complete with a breakdown of those flaws? The more popular the game, the better.
- Harvest Moon
- Animal Crossing
- Dance Dance Revolution
- SimAnything
- Sports games
"You cannot find out which view is the right one by science in the ordinary sense." - C.S. Lewis on Intelligent Design
Candy Cruncher is actually a really fun game. Not th e sort of thing I could play for a couple hours like say quake, but fun. My preferred version of it is the http://eongames.com port to Qtopia.
The windows version is coded for DirectX/Direct3D or whatever. The Mac version is coded for Carbon. Both of those are native Windows APIs. Their Linux port is made in SDL, and is probably statically linked, as opposed to the other two.
Karma: Could be worse (could be raining)
If you design your plug-in system well, it isn't surprising that you could write plugs for various API to make the most out of them.
A normal Windows PC game would go like this: Game Engine -> Direct X -> Display Driver -> Graphics Card Internal Code
A plug-in system would be: Game Engine -> Plug-in (DirectX,OpenGL,etc) -> 3D API (DirectX,OpenGL,etc) -> Display Drivers -> Graphics Card Internal Code
This way when you are porting to another OS you don't have to worry about all the features your using be supported by OpenGL patches. With this system it would be possible to develope an engine that contained output plug-ins for the Windows (DirectX), Linux(OpenGL), OSX(OpenGL), PS2, Xbox, and Gamecube. The only place I see issues with running it on the PS2 due to having to stream textures rather than loading everything into video RAM.
Why, you may ask this isn't done more? It's Simple. Most game are designed for a target platform and coded directly for it. Later when it is decided to have the game go multi-platform, then the issues appear. If you plan on going multi-platform from the start you can design more your routines more efficiently.
Why not just plan to make all games multi-platform? Developing for multiple platforms is more difficult and takes more time and money. It's alot easier to say "I'm creating a Windows game" and then just have your engine output to DirectX. Later when you want to port to Linux or OSX and need to use OpenGL, you run into the difficulties.
From what I've heard, he's been going through his entire life as a BS artist. He's been been brought up in converstation as an example of the marketroid-wanna be programmer type to be wary of. I don't know these as true first hand, but from what I've heard, in school he cheated, at 3dfx he took credit for other peoples' work. At Id, they realized their mistake and fired him - same thing at verant.
Also for your information Ryan "icculus" Gordon did the initial SDL porting and I am the current maintainer of the SDL as well as Qtopia ports.
I don't believe it's sour grapes at all, you bring up a lot of very valid points. I was primarily known for having a big mouth and a lot of opinions, and those aren't the types of things you base a career on.
In fact, in my defense, if you look at publicity for "Brian Hook", there has been none, because I've reached the point where building something on my own is more important than self-aggrandizement and promoting a name or personality.
At 3Dfx I was ancillary -- and, just to clarify, while I was a major influence on Glide, the Glide that everyone knew was actually significantly cleaned up and fixed by people after I left 3Dfx, so associating it strictly with me is a disservice. 3Dfx was about the hardware.
At id, I came in late on Quake 2 and left early on Quake 3, and in hindsight, it was a blown opportunity for me because I didn't contribute as much as I could have. But then again, trying to code around John Carmack is fairly demoralizing because he makes everything hard seem too easy.
I consented to this interview -- my first in THREE YEARS -- because I figured CG was a small enough site that no one would really notice. The questions that were asked I thought were extremely good and insightful and, frankly, I wanted to slowly start building awareness within the Linux community as we start pushing out more products.
But anyway, I think you're pretty accurate on your assessment, but I feel obligated to post just so you (and others) understand that this isn't a typical washed up, pathetic developer who has lost his relevance looking for one last moment in the sun.
I fully intend to build up Pyrogon into something important and fun. This is what is most important to me at this point, because the quasi-fame I had before was fleeting and, in the end, irrelevant to my own enjoyment of developing games.
And when Pyrogon is successful, no one will be able to make a post like yours saying it wasn't earned, since only two people work at Pyrogon, and it's 100% self-funded. It lives or dies by our abilities.
Of course, until Pyrogon has reached that stage, comments like yours will continue to be made, as they probably should.
-Hook
Stellar Deep really does require a certain amount of financial and time committment we can't justify at this point. We're not particularly interested in finding outside sources of income for this, since the whole point of being independent is so we can pick and choose what we work on.
So with that in mind, we're trying to build up a revenue base so that we can take 6-12 months to sit down and do something with Stellar Deep. It was originally mentioned on the Pyrogon site when we launched because it was a "Wouldn't it be cool if?" type of thing. Unfortunately it came off as a product announcement, and I hate vaporware, so we pulled it but kept it in the forums.
We think it'll be a fun game, with all the aspects of Elite, EV:Nova, Subspace, et. al. that people really enjoyed, and hopefully with an added social and strategy dynamic lacking in those games.
One day...
-Hook