Buggery! Games Makers Reject New Law

you took the words right out of my mouth and add detail congratulations

As a programmer in the games industry, I can say with certainty that this simply could not work. As projects get bigger and more complicated, problems will creep in that are hard to detect. Even a team of 50 testers working on a game solid for a year could not find all the bugs that the public is likely to find in a month after a title goes on sale.

It is not possible to offer guarantees that software is bug free, all you can do is make a reasonable effort to ensure that it is stable under normal circumstances.

Perhaps Public Beta Testing may become more important, as COD and Halo 3 did maybe more single player experinces need a beta testing service.

This is another dumb law, made up buy people who don't understand how games are made and work. They see all tech as just that, when there's several different thing can be created out of code. To label them all with such a sweeping label would be quite mad.

Even an OS like Windows has it problems, because creating stuff from one and zeros is not a a perfect science. People are prone to error, just as much as what they create.

See, I told you the EU are trying to build a Utopia of impossible nazi like perfection.

Stuff you, if I have to download a patch to get it to work properly it is defective.

Windows has problems because of the way it was designed and programmed, by humans shock horror. There are some hardware issues that are not constantly/or within spec that needs to be addressed, that even most games programmers int he industry might not be fully aware of. However, if you don't strain to get the last 20% performance wrenched out of the system you might find 80% less of those problems. How much will that hold back the "creativity" next year the machines will probably run 20% faster.

Standardise API abstraction (/hardware) and game engines, it might loose you 20%+ but that would help coding eventually too. Now it is up to the programmers and designers to nail down development (and before any "industry" expert chips in, realise how few good programmers there are), after all, game development should be about design and content, not the technology.

@Ste

Wow! I've never heard such an un-educated bunch of crap since I last listened to a politician.

It is simply not possible to standardise API abstraction in games consoles since all the major 3 systems work in completely different ways. At a very simple level, look at the controllers - each system has something unique to their controllers whether is be audio feedback, number/layout of buttons, rumble. If you try to standardise it all, you just basically end up with the lowest common denominator from all the systems or your code simply won't work, so you'd lose analoge stick input, motion sensing, rumble, audio feedback and a ton of buttons. How's that help anything?

Our studio has a common API/engine that runs on Wii, DS, iPhone, PSP with PS3 and 360 support being added. Despite this, there still has to be tons of specific "ifdef PLATFORM_WII" code or it simply will not work.

Bugs are not caused by the programmers trying to get the most out of the hardware, they are simply unforseen clashes between two lines of code. If you have a game with 1000 lines of code, it is easier to keep bug free than if you have a game with 1000000 lines of code. Simple maths.

If this did come into power it should only be for bugs that prevent you progressing in the game, if a patch can't be obtained (e.g. the user doesn't have net access) then the user should be entitled to their money back, as in the case of the Hospital Tycoon game mentioned. My misses had a similar situation with Sims 2 on the DS and too many aliens in her pockets or something!
Proving a fault (especially on PC) could be the tricky part.

I don't have a problem with having to download a patch for a game, mistake gets made or things get over looked, it's fact of life. Even cars get recalled and "patched"!

So making developers/publishers adhere to this for 2 years might not be a bad thing.

software is never bug free, and every software developer knows this.

HOWEVER they also know that when a bug is reported/detected it is assigned a priority:
1) it crashes my machine, and reformats my hard disk every time it starts up.
...
5) Why doesn't it make tea for me after I've completed a level.

or something similar.

You're never going to get them all - some are just too trivial to risk introducing more serious bugs, and the cost is prohibiituve, but any "normal" software vendor offers patches/updates to fix the biggies. As most consoles are now online, it's really not unreasonable to expect fixes to MAJOR bugs within the first few years of a games life. Simlarly "support" would cover mainting support for online play for 2 years - again not unreasonable.

Trading standards dont; expect tangible products to be "perfect', just functional to a reasonable standard. Games should be no different.

@Joji
read the source article instead of looking dumber then your accusations, it says the potential law is aimed at games with glitches that prevent users from finishing the games. Like Haritori's first example. Also, the law hasn't been proposed yet to the EU parlement, it's just the commission head saying they're looking into it and talking with the industry in question to come to an agreement. And they're allready accounting for possible abuse from users. The end form is probably going to be riddled with special exemptions for digital goods in the end.

And what the crap is Nazi like perfection when they're just talking about a possible 2 year garantee for all products... So toyota is even more nazi with its 5 year garantees? And the nazi's were hardly aiming for perfection. Nazi-germany was quite ineffeciently organised with party-officials meddling with civil, industrial and research officials and back. Hitler never allowed the two structures to merge because this would create potential consolidated power-positions that could challenge his position. It's just that german efficiency in general is so high, nobody outside germany noticed this.

@h

You can never make software bug free, end of story. How would you even define what is a bug and what isn't? Is it a real bug or a feature request? Who decides that? A lawyer who gets his secretary to check his email for him? As stated earlier by someone the best you can hope for is that a reasonable effort is put into making it as stable as possible.

You try to make games (or any software) completely bug free you could multiply the development time (and therefore cost) by 10 and you'll still end up with something with bugs in it.

@irritant

You are acting like an idiot, if you want it even more factual, with your head up your ass, a petty know nothing really, sub average (which equals average in the PC world) programmer by the looks of it, and now you known what you can do with your Semi-educated opinion. Amazing the morons around here, who think they even know computer science.

Practice and experience at solely being a moron tends to make a moron a bigger moron, not a good opinion maker.

I simply have to go on.



Exactly! You never considered writing a self configuring API that will remap to different controllers, even ones yet to be invented, in an optimal configuration, and accept preferred configurations put in by developers (data configuration file), well I have laid out such an API. Likely because you are a naive programmer depending on API writers to do all the tricky hard work for you, not realising just how much can be done. You are already using an API to abstract away from problems, usually sacrificing some performance (compared to low level programming) and reducing bugs, as I was talking about. You will notice I ws referring to the 80% and 20% in relation to Windows/PC cutting edge programming as a off the head remark, then go to expand this example to consoles and standardised systems (consoles which have less variations over time to cause problems, which i should have considered). So it is hardly worth going on and saying Java (as not so good example), or

http://www.khronos.org/opencl/ (which is worth reading for those differing hardware resources)
http://www.khronos.org/opengl/ (the path forward aims for more compatibility and getting rid of outdated structures, that it has desperately suffered from in times past).
http://www.khronos.org/openvg/
(was also an open sound API)

Not to mention the others popping up:
http://www.khronos.org/openmax/
http://www.khronos.org/openml/
http://www.khronos.org/openkode/

Or the number of attempts to do what I was talking about.




Of course, but I never said your API was excellent or what I was referring to, even API designers have different levels of abilities to one another.



Depends on the market. Referring back to PC's the different hardwares, and trying to drive it to the extremes produces coding problems because of the massive variations between hardware and some in between revisions of the same hardware. You may notice that fixes to Windows, DirectX, graphics card drivers, other drivers and bios in total happen often, and you notice in the graphic card drivers revisions that fix bugs in various games. In the console scene, the hardware is standardised from system revision to system revision (usually, concerning game play), and over time the bugs are understood and fixed, even carried over if it maintains compatibility) and hidden behind drivers, firmware fixes and API's. If you press past this safety cushioning to press the system people have been prone to find all sorts of unexpected behaviour (which cutting edge performance techy oriented games have previously had a habit of doing in the PC world in times past) but likely much less than the PC. It is the firmware, drivers, API's and cross compatible API's that are protecting you from much of this, but that does not mean that they are excellent in the way they try to do it. But as I referred to before, you are already cushioned and doing this.

You will notice I identified a number of things that caused bugs, and strategies to reduce them, if you already are innocently following some of these strategies you will see a different picture. Simple programmer's mistakes are also big issues, and careful diligence and intelligence is needed (not a trait so enjoyed in the PC world). But you will notice that in my example I stated that I was referring to Windows (PC) issues, and that the large human problem in coding, which you have illustrated.

You would be surprised at just how much politicians know when they are not trying to lie to you.


I apologise for my outburst before irritant, but I have come across a number of such things over the years, and get tired of it.

Now I am awake.

Actually that is not right, it is just that for big projects, or stressed, hurried or slack programmers and designers etc, things get through, which is the standard experience in the PC world. In the classic real time embedded/consumer electronics world, of real programmers, programs can be a lot smaller and requirements a lot higher. I am pretty sure (in this age of bigger embedded programs, even using windows) you will still find examples of no bugs at release.

The use of a standardised game engine allows bugs to be worked out over revisions of the engine that can be then used for future game, or patched to the existing game, leaving just the custom code, action intelligence, content etc to define. API's on their own go a little way to this.

Im no programmer or going to try and understand this, but if a standard API exsisted wouldnt that API have bugs which the programmers would have to correct a possibly missing some of those errors before retail, therefore making the point of it mute?

like i say i dont understand programming, in fact my limits are a few lines of html, and tinkering with my autoexec.bat & config.sys back in the day, but i refuse to believe standardising programming specifics would eliminate glitches and bugs.

something has to be done even if it's a mandatory minimum of 1 year support, I don't know about you guys but if I come across a bug that prevents me from completing a game, I'm either going to want it patched or my money back.

@h doesn't that depend on the nature of the patch?

if it's just for gameplay enhancements, say to improve bot AI or put you in a better spawn point, then I wouldn't call that defective.

if it;s fix a minor annoyance, like bots getting stuck on scenery then yeah it's defective but not worthy of "money back now".

crash or game progression bug = "money back now", imo

does nobody honor godwin's law any longer? pfft!
(cf. optimusp)

Not so much. Ideally the API programmers know their stuff, do the code carefully and well, so that it does not have to be repeated in thousands of programs by other programmers, and it is done so it can be simply and efficiently used. As the API is reused (and maybe part of the system, like DirectX) it can be cleaned up over time. It was meant to manage the quality of code to, ideally reducing bugs and reducing them over time. But in the real world things often do not work that way, the stress of high output and low quality programmers gets in the way. The difference between the worse and best programmers is something like 1000 times, not 1000 % (10 times) but 1000. If you find a 1000x programmer you stick him in a room call him "House" and let him go. You will probably find few 100x programmers, and a ten times programmer is probably still not that crash hot. Unfortunately industry has not nailed down this work division for us, where the really good programmers take on the quality load for the rest of us. by doing most of the work in firmwares, drivers, abstraction layers, API's, middle ware and libraries etc. The really good analysts and designers need to make things simple and efficient too. In a gaming OS you would spend a lot of time defining the interfaces and functions with a mind to how the code will work. In a perfect system it would be perfectly bug free, and the application programmers merely call functions and design some control logic, even logic, movement and actions could be defined in data files, I do not know of such a system (though some authorware can come close to the last bit, but are often weak in ability).

When some people are stuck with research essay creating.

[25 Feb 2010, 13:45: Message edited by 'tyrion']
No Spam! Ta!