A very good GDC session

[passerby]

Posted up at a few forums. Very good read so I'll link it here too. Long but really worth reading all the way to the end - especially since it drops all the PR BS.
http://crystaltips.typepad.com/wonderland/2005/03/burn_the_house_.html

 

[Joe DeFuria]

Wow...what a bunch of whiners...

 

[Edge]

> " Has everyone bought Bioware’s online modules? JUST BUY THEM, OK, even if you don’t have the original games!"

That's Warren Spector saying that. Is he trying to be funny, by making his point?

I stopped reading after that.

 

[ERP]

> " Has everyone bought Bioware’s online modules? JUST BUY THEM, OK, even if you don’t have the original games!"

That's Warren Spector saying that. Is he trying to be funny, by making his point?

I stopped reading after that.

The way I read it FWIW

He's being funny, he's saying that Bioware tried a different form of distribution and it hasn't worked aswell as they liked and it would be good for the industry (It needs to happen) if someone was successful at it.

 

[ZoinKs!]

He's being funny, he's saying that Bioware tried a different form of distribution and it hasn't worked aswell as they liked and it would be good for the industry (It needs to happen) if someone was successful at it.

You're partly right. He's saying it's good to develop distribution methods other then cardboard boxes at retail stores, so he wants Bioware's Premium Mods to succeed.

Here's where you're wrong: Bioware's downloadable modules have been successful. More development has been greenlighted. It's even expanding beyond the original scope: I know of two under development by non-Bioware community members (one by Stefan Gagne and one by the Dragonlance Adventures mod team).

Profits from Premium Modules are the source of funding for all further nwn patches. They say more people are being paid to work on nwn now then at any time since spring 2002.

I'll also put in a plug for the Prem. Mods: I bought Shadowguard and Kingmaker and consider both well worth the money. Maybe one of these days I'll start a topic on them...

---

Now, re. the linked article. I'll echo passerby: It's an excellent read... quite thought provoking. These people are more passionate about gaming then the most hard core internet forum freaks.

And it'd definately worth reading all the way to the end--- like I said, thought provoking.

 

[jvd]

I would have been heavly interested in nwns if it had worked properly out of the box . These models for distrubution are great but the original content has to be there . Hopefully they wont scarfice one for the other

 

[darkblu]

i particularly liked that guy's rant about how bad the average spaghetti code out there runs on in-order cpus.

well, life sucks, and then you die. aside from that, some people need to realize that the fact that a knife makes for a bad fork does not make the knife a bad tool, it's just not a fork. but hey, if someone wants to use their knife as a fork - all power to them.

 

[Johnny Awesome]

This session was just sour grapes from old has-beens.

 

[passerby]

Well judging from reactions around forums it appears that a lot of things have been said that the fanbois don't like to hear.

But that's ok, I just enjoy being devil's advocate once in a while.

I'm interested in more opinions about the disadvantage of in-order-CPUs mentioned. Any personal experiences from anyone? Is it a hugh problem? After all initial Xenon dev kits are using out-of-order PPC CPUs, so any shocks when you received the actual CPUs and the logic performance suddenly nose-dived...?

 

[darkblu]

I'm interested in more opinions about the disadvantage of in-order-CPUs mentioned. Any personal experiences from anyone? Is it a hugh problem?

i think one could extrapolate from other in-order devices. e.g. the original pentium (p5) architecture, which was superscalar (two pipe), in-order and had pretty rigid restrictions re what goes where and when. of course, at that time the memory gap wasn't as wide as nowadays, so cache-miss pipe bubbles were of a different magnitude.

After all initial Xenon dev kits are using out-of-order PPC CPUs, so any shocks when you received the actual CPUs and the logic performance suddenly nose-dived...?

if that's the case (as we don't know for sure) then it could positevely be a problem for getting any remotely-true performance metrics for the code. which could bring up the question of what did ms think sending out dual oo machines to emulate a tripple in-order platform. aside from being another give-away gesture, that is.

 

[ERP]

Alpha kits aren't for performance metrics, they are there to let you work. They have the same API's, some features are emulated, others are missing. You can work on your graphics, get your pipelines in place, write your game code.

But...

Anyone who spends any significant amount of time optimising on an alpha kit is wasting their time. I think there is performance info you can draw from the alpha kits (I saw an interesting performance anomoly last week where a compiler optimisation caused a 30x performance drop, it would do the same thing on the final hardware), but for the most part you should save the optimisations until the Beta Kits show up.

 

[ERP]

OK I have to ask the people who view the comments as "moaning" why you read it that way?

For the most part I thought the panel members were really just identifying where they feel the industry needed to change in order to be more open to creative game design. Or to raise the quality bar...

 

[Joe DeFuria]

OK I have to ask the people who view the comments as "moaning" why you read it that way?

Because I read it as a bunch of people wanting to reap some rewards (total development freedom, more profits for self), without paying much attention to the risks involved in doing so.

For the most part I thought the panel members were really just identifying where they feel the industry needed to change in order to be more open to creative game design. Or to raise the quality bar...

I agree...unfortunately, the "vibe" I get is like:

"Yeah! Baby! This is what we need to do! Yeeehaaa!! Solidarity! Fight the Man!! Change the world!!"

"Now which one of you wants to actually stick their neck out and do it...it ain't going to be me....I'm um...too busy..."

 

[Lumine]

I agree with Joe Defuria in that it does seem to be alot of "ra! ra!" and not as much actual plan, however I don't see that as something that detracts from the value of what was said. A number of the speakers mentioned the need for smaller budget/independent games to be made. What sort of things would help this along? What sort of format would be used at an independent game festival? What sort of tools are needed?

 

[ERP]

I know a few of the people in that talk and I'll agree there is an element of "screw the man", but a lot of that comes from having seen the industry evolve. <10 years ago it was still possible to be a small developer producing AAA titles, and building games you wanted to build.

FWIW I'm not sure there is an answer.

But to me the most licid point was the decoupling of Publishing from funding. Publishers want what they can sell to Wallmart, they are risk adverse for good reason. While funding and publishing are tied together you basically have Wallmart (albeit indirectly) and not the creative people designing your games.

The biggest urdle to "indie games" is money, what you can do with 2-300K these days is so limited. And I can't imagine anyone ponying up 2+ Million on an untested developer/product.

Perhaps if there was another cheap effective form of distribution (perhaps an online channel), we might see 3rd parties funding development and start to see more diversity in product. But if that doesn't happen for another 5 years I have to wonder what the cost of entry for these "3rd parties" would be.

 

[j^aws]

...FWIW I'm not sure there is an answer.
...

I always thought that Sonys Net Yaroze initiative for the PS1 would kickstart this indie/ bedroom coding scene. Apart from some demos distribute on magazine covers it really hasn't taken off to what I imagined because of the current distribution model.

Maybe the PS3 yaroze/linux kit equivalent and Sonys broadband talk and a rumoured micropayments model could make this more realistic. Also similar initiatives from MS/ Nintendo perhaps...

 

[ERP]

...FWIW I'm not sure there is an answer.
...

I always thought that Sonys Net Yaroze initiative for the PS1 would kickstart this indie/ bedroom coding scene. Apart from some demos distribute on magazine covers it really hasn't taken off to what I imagined because of the current distribution model.

Maybe the PS3 yaroze/linux kit equivalent and Sonys broadband talk and a rumoured micropayments model could make this more realistic. Also similar initiatives from MS/ Nintendo perhaps...

I hate to say it, but the days of building a AAA game in your bedroom have long gone. About the best a program like Yaroze can offer is preparing people to work at larger development companies.

Even building a high quality demo to demonstrate the play mechanics and look your going for to a publisher is an enormous undertaking.

To give you some idea the player character in a game I'm working on uses 100's of animations just for the basic play mechanics. The investment in Time or money is just so prohibitively high that someone in their bedroom simpy can't do it any more.

For point of reference, the first published game I wrote (in fact in my bedroom) the player had exactly 1 animation (walk, jump was one of the walk frames).

 

[Shifty Geezer]

One of the huge limiting factors has to be the licensing model from the console manufacturers. Paying up front for printing of games, regardless of whether you sell them or not, limits IP to proven franchises. It's not just the development funding you lose but the licensing cash as well.

A shift to online distribution where licensing is on units sold would be a big improvement. And though we won't see AAA titles similar to existing games, we could see AAA original ideas, using simple concepts that aren;t character based. eg. Something like Mercury or something using geometric objects in a puzzler would need simpler art assests than a massive RPG or FPS, yet image quality and game dynamics could be on a par with big name titles, especially using existing libraries (physics for example) to reduce the coding requirements.

 

[Laa-Yosh]

Warren's always been a hero of mine... what was his first game, Space Rouge? That came out on the C64 in the eighties, dammit...

 

[j^aws]

...FWIW I'm not sure there is an answer.
...

I always thought that Sonys Net Yaroze initiative for the PS1 would kickstart this indie/ bedroom coding scene. Apart from some demos distribute on magazine covers it really hasn't taken off to what I imagined because of the current distribution model.

Maybe the PS3 yaroze/linux kit equivalent and Sonys broadband talk and a rumoured micropayments model could make this more realistic. Also similar initiatives from MS/ Nintendo perhaps...

I hate to say it, but the days of building a AAA game in your bedroom have long gone. About the best a program like Yaroze can offer is preparing people to work at larger development companies.

Even building a high quality demo to demonstrate the play mechanics and look your going for to a publisher is an enormous undertaking.

To give you some idea the player character in a game I'm working on uses 100's of animations just for the basic play mechanics. The investment in Time or money is just so prohibitively high that someone in their bedroom simpy can't do it any more.

For point of reference, the first published game I wrote (in fact in my bedroom) the player had exactly 1 animation (walk, jump was one of the walk frames).

I know where you're coming from. They only way for change is for alternate/ cheaper distribution models and the only realistic alternative is online. Since all three players are making a fuss over online next gen, it's a matter for these players to setup these infrastructures to help indie/bedroom coders. They don't have to be AAA games nor do AAA games have to be big budget. One of my favourite games of all-time is Stargate aka Defender 2 (arcade/MAME) in all it's 16 colour glory! ...And for indie/freeware stuff on the Mac OSX platform, I came across some nice looking stuff here,

http://shinh.skr.jp/osxbin/

Now all we need is to be able to download games onto our consoles with some sort of micro-payments model. Budget game prices or try before you buy. But Sony/MS/Nintendo would have to create this infrastructure with the publishers...or we're really stuck with the status quo...