next »

[ThemsAllTook]

For instance, if your CPU is too slow for good AI, then just make AI worse on that computer. It's better to play a slightly worse game than not be able to play at all.

I agree with you in principle, but I really can't see this specific example as a good idea. If you change AI behavior in any way, you're playing a different game. If you're doing it based on real-time performance, you make your game nondeterministic. If you do it by querying the CPU characteristics...that's another can of worms in terms of hardware support. As the developer, you'll have to test and fully support both (?) cases as well as the switchover mechanism, which introduces a potentially huge amount of additional work for you. Even if you really want to go that much trouble, it seems game-breaking in too many ways to be worth consideration. Playing a broken game probably isn't better than not being able to play at all.

I'd find it more appropriate to 1) use an AI implementation that's likely to be fast enough on any computer that can reasonably be expected to run my game (say, anything made in the last 5 years), and 2) just let it be slow if it's slow on a low-end CPU. A slow game is better than a broken game.

[Edit: Yikes, veering quite a bit off topic here...hope this doesn't cause a derail]

[Danmark]

I've seen many games have minimum resolution requirements displayed in the system req's, especially for older games.

Me too. I was oblique about it, but what I meant is that it's reasonable for resolution to be part of the abstract minimum requirements, not necessarily the ones provided the developers. Of course developers should provide good estimates/true values of any requirements, which is another matter.

Speaking of which, how common is it for fairly high resolution requirements (such as ones too high for your netbook) in contemporary games to be clearly stated?

For instance, if your CPU is too slow for good AI, then just make AI worse on that computer. It's better to play a slightly worse game than not be able to play at all.

Couldn't disagree more. As ThemsAllTook explains, it prohibits developers from working on just a single game per game project, which I think is a crucial requirement. Quality would certainly suffer. It's bad enough that developers can't enforce good reproduction of sound and image on the end-user's speakers/headphones or monitor respectively. Best to control as many aspects as possible of a game being played.

[Vino]

For instance, if your CPU is too slow for good AI, then just make AI worse on that computer. It's better to play a slightly worse game than not be able to play at all.

Maintaining low spec functionality is a huge pain. It's hardly as easy as you make it sound. And then, some people are going to blame the shitty AI/graphics/gameplay on the developer or the game and form their impression based on that. Many developers would rather people not buy or play their game than play a gimped, cut down version that won't be fun for anybody.

If you have an old computer, don't expect to be able to play every new game. Simple as that.

[_Tommo_]

Maintaining low spec functionality is a huge pain. It's hardly as easy as you make it sound. And then, some people are going to blame the shitty AI/graphics/gameplay on the developer or the game and form their impression based on that. Many developers would rather people not buy or play their game than play a gimped, cut down version that won't be fun for anybody.

If you have an old computer, don't expect to be able to play every new game. Simple as that.

this.
And I really wonder anyway what kind of AI you can have that takes more than 1% of frame time even on an old pc. It has to be something massive

[Jasmine]

I really wonder anyway what kind of AI you can have that takes more than 1% of frame time even on an old pc. It has to be something massive

Don't forget that
- some modern languages can be slow and memory hungry compared with C/assembler. With such modern languages, the AI doesn't have to be that massive before it starts impacting the framerate.

- programmers nowadays can get away with writing sloppy code, knowing that modern hardware can pick up most of the slack. A sloppily written AI can quickly use up that slack. If you develop on an older-spec system, you will learn to write efficient code. But if you develop on a top-notch system, you won't necessarily know if your code is sluggish or not.

[Klaim]

Strategy games might require strong coding and performent processing when the AI is a "bot".

[_Tommo_]

But then, if the AI is sluggishly coded you don't need to downgrade it, you just need to optimize. If it's inherently heavy, I don't see how you could scrap parts of it without damaging your game's playability.
Imo you just have to set a minimum and assume it is always there, trying to scale on everything is pointless.

[nospoon]

AI is usually a simple state machine, and it doesn't eat too much cpu.
The pathfinding usually eats the most cpu time in AI, but there is alot room for optimization in it.

[Klaim]

_Tommo_ > I fully agree with you.

AI is usually a simple state machine, and it doesn't eat too much cpu.
The pathfinding usually eats the most cpu time in AI, but there is alot room for optimization in it.

Let's not generalize, I've personnaly wrote some different kind of AI and some kind of games does require optimization on pathfinding AND general AI code (strategy games, management games, tycoons etc). Not every game run with a state machine, even indie games.

[Core Xii]

Couldn't disagree more. As ThemsAllTook explains, it prohibits developers from working on just a single game per game project, which I think is a crucial requirement. Quality would certainly suffer. It's bad enough that developers can't enforce good reproduction of sound and image on the end-user's speakers/headphones or monitor respectively. Best to control as many aspects as possible of a game being played.

Maintaining low spec functionality is a huge pain. It's hardly as easy as you make it sound. And then, some people are going to blame the shitty AI/graphics/gameplay on the developer or the game and form their impression based on that. Many developers would rather people not buy or play their game than play a gimped, cut down version that won't be fun for anybody.

If you have an old computer, don't expect to be able to play every new game. Simple as that.

Coming from a web development background, I find these views appalling. All websites should work on all browsers (with a degraded experience if necessary), and I apply the same philosophy to game development. You can't control every aspect; therefore you shouldn't attempt to control any of them. Exercise fluid, flexible design. Make games that scale, up, down and sideways if necessary. Identify the core of your game, then make it build up as more resources are available (bigger maps! smarter AI! more special effects!). Is it more work? Of course! That's how art is made. You don't take the easy road and just assume everyone has the latest and greatest in computing or target one specific machine (like console devs do... look at the crappy PC ports we get).

[iffi]

Couldn't disagree more. As ThemsAllTook explains, it prohibits developers from working on just a single game per game project, which I think is a crucial requirement. Quality would certainly suffer. It's bad enough that developers can't enforce good reproduction of sound and image on the end-user's speakers/headphones or monitor respectively. Best to control as many aspects as possible of a game being played.

Maintaining low spec functionality is a huge pain. It's hardly as easy as you make it sound. And then, some people are going to blame the shitty AI/graphics/gameplay on the developer or the game and form their impression based on that. Many developers would rather people not buy or play their game than play a gimped, cut down version that won't be fun for anybody.

If you have an old computer, don't expect to be able to play every new game. Simple as that.

Coming from a web development background, I find these views appalling. All websites should work on all browsers (with a degraded experience if necessary), and I apply the same philosophy to game development. You can't control every aspect; therefore you shouldn't attempt to control any of them. Exercise fluid, flexible design. Make games that scale, up, down and sideways if necessary. Identify the core of your game, then make it build up as more resources are available (bigger maps! smarter AI! more special effects!). Is it more work? Of course! That's how art is made. You don't take the easy road and just assume everyone has the latest and greatest in computing or target one specific machine (like console devs do... look at the crappy PC ports we get).

IMO in most games the AI would count as part of the "core" of the game. I wouldn't want enemies (or allies, for that matter) to behave differently because of something as unpredictable as system specs. I would limit myself to downgrading the presentation of the game (graphics, sound, etc.) rather than the game itself and its mechanics (AI included). And yes, what constitutes presentation vs. mechanics is debatable.

Besides, it's not like you can get every game to work with every computer. There's usually some lower limit of how far you can degrade a game before it gets impractical to go any further. No matter how primitive you make Battlefield 3's graphics, AI, sound, etc. it's not going to run on an NES-tier system without becoming a fundamentally different game. Support as many different systems as you can, but you have to draw the line somewhere. As for "All websites should work on all browsers," I personally don't see what's wrong with not bothering to make your website compatible with IE 1 or something like that.

[Vino]

Exercise fluid, flexible design. Make games that scale, up, down and sideways if necessary. Identify the core of your game, then make it build up as more resources are available (bigger maps! smarter AI! more special effects!).

To me this sounds like you volunteering someone else to do a lot of work.

The world of web browsers and the world of desktops are two different things. The same paradigms don't apply. You can expect most websites to work on old browsers, but you can't expect the newest games to work on old hardware. Maybe large companies can handle this because they can assign people specifically to writing and testing code for old hardware. Indie developers don't have that luxury - my time is precious and every minute I spend on old hardware (assembling, writing, testing) is a minute I don't spend on making a fun game.

Seriously, who has a six year old graphics card lying around? Do you really expect developers to make their game run on an ATI Radeon 9600? In some cases this requires maintaining two or more different rendering pipelines. No thanks.

[Jasmine]

At one time, computers used standard resolutions. For example, with the Amiga-500, 320x200 was the standard, and everyone who used that machine was expected to run it at that resolution.

Some people no doubt look down on this, but standards do bring some benefits:

(i) developers can develop for a particular resolution, confident in the knowledge that all users will be using that resolution.
(ii) users can use that resolution, confident in the knowledge that all applications will be designed for it.

In these modern times, there is so much variety of monitors and resolutions that standards have become eroded.

So now users can never be sure that an application will appear as intended on their monitor's native resolution. While developers have absolutely no idea which of the myriad resolutions to cater for, so graphics has to be shoehorned into place, causing sharpness to be lost to interpolation. 

Standards are valuable.

[iffi]

<snip>

This becomes painfully apparent whenever I find a wallpaper that isn't available in my screen resolution. You'd expect 1920x1080 to be a standard people can actually agree on, but no, I still have to crop a lot of 1920x1200 wallpapers...

[Cironian]

This becomes painfully apparent whenever I find a wallpaper that isn't available in my screen resolution. You'd expect 1920x1080 to be a standard people can actually agree on, but no, I still have to crop a lot of 1920x1200 wallpapers...

As far as I'm aware 16:10 (as in 1920x1200) is much more common as the native resolution for computer monitors, while 16:9 (1920x1080) is the standard for other displays like TVs and movie projectors. I don't expect those two fields to agree on the one true aspect ratio any time soon.