Yes, it very much is.

I've never had a problem with anything on the PC not working (except for hardware).

BS. I've been programming C++ professionally for near on 8 years now. It's no secret why so much software comes out buggy as hell, and it isn't some nonsense about coding being "as much an art as it is a science."

1: Arbitrary release dates. Particularly those that don't actually take coding time into account. As a corollary, most programmers can't estimate the time it takes to do tasks very well, particularly those that they've never done before.

2: Programmers who do not write good code. Some programmers write bad code because of #1; they weren't given the time to do it right. Some programmers are uneducated, and I don't mean lacking a college degree. I mean they don't know how to even recognize good code, let alone write it. Others were "educated stupid". That is, they think they know what good code is, but clearly do not. Born-again template or OOP or whatever paradigm-of-the-week coders fall victim to this.

3: No testing. Most software products don't even do basic testing of the code they write.

4: Apathy. They can always patch later. Right? Why bother testing and so forth when you can just ship crap and fix it later?

In this day and age, with all the various tools and libraries available to programmers, memory leaks should not be acceptable. Yet GalCiv2 shipped with a memory sieve. Why? Probably a combination of 1, 3, and 4.

Writing good code won't guarantee bug-free code. Especially for things like games, where the very game design can be buggy; that is, even if you implement what the designer says to do, you have a bug because what the designer said was wrong. There will always be that kind of bug. But any form of crash bug or memory leak or the like is totally unacceptable in this day and age. Good coding schemes may not guarantee the lack of "feature bugs" (feature X is stupid or feature X doesn't work), but it does guarantee (or close enough) the lack of crash bugs and the like. That is, clear programmer errors.



Until those factors are fixed, PCs will be a niche part of the larger videogame industry.

Call me, I'll test anything with modest sys-req's.

Totally agreed. The most buggy projects are the rushed products.



There are good and bard parts of OOP. If used properly, it's a very good way to structure software. If used improperly, however, the overhead and poorly thought out dependencies can kill code.



Totally agreed, although that often goes back to #1. They don't test because they're on a time schedule.



Agreed, although it's a lot easier to make C# or Java leak free than C++. In addition, it can very well depend on the quality of the libraries you are using. I've seen my share of very poorly constructed tools and libraries.

. . . and not all "bugs" are in your own software. When developing for an architecture as wildly varied as the x86, there are plenty of hardware, OS, and driver bugs you may encounter and need to work around. This is especially true in game software, which demands a lot more driver level and lower level access than your average application.

It can be difficult to make your own software solid when the libraries and drivers it depends on have bugs.



I don't think that is necessarily true. Just because you're on a personal crusade against buggy software doesn't mean that people have stopped buying games because of bugs. I'm not saying your crusade is misdirected in any way, but how much is it affecting sales, and is there a good solution for the issues you've raised?

It's nice that you've found a problem to complain about. A proposed solution would be better. An implemented solution is best.

I've always thought that was a given, in anything, not just gaming.

Of course it's hard to build around bugs, but that's not really your fault as a developer, is it?

No, it is not. Well, not in the real sense of the program using less memory. Garbage collection doesn't fix the overall problem of the program using more memory than it needs to. It simply helps ensure that the memory usage problem won't get out of hand.



Yes, but there are free ones (in both senses of the word) that are readily available to any C++ programmer on any platform. There's no excuse for not using them.



It is a fundamental problem of the PC platform; there's no getting around that. The lack of which is therefore a fundamental advantage of console games. No matter how poorly made a console game is, it will run or crash based solely on the code written by the developers.



Why are people so afraid of the word "niche"? People act like calling something niche means that it's insignificant or something. Yes, people are not going to stop buying PC games because of bugs. But PC games are inherently more buggy than console games. And that's the major reason why PC games are, and will remain, a niche market compared to console games.

It's a niche you can certainly live on. But it will never be bigger than consoles.

Actually, yes in the case of Java. A Java application can only allocate so much memory, and the user has to explicitly run the JVM with certain command line parameters to allow it to use more memory. A memory leak is not allowed to continue indefinitely in Java. And, of course, the whole thing is always cleared when the program terminates.



A library doesn't have to be commercial to be poor quality. The quality of the library depends on the developers who maintain it, not whether or not it is open source. In addition, what libraries I use also depends on what I want my software to do.



Consoles don't lack bugs: It's just that after the first year or two, most of the bugs are known and pretty much all of the developers and libraries have built around them. You will find this effect on Macs, and it has shown up very clearly in the XP vs Vista debates: XP has had a very long time to wrinkle out and work around all of the bugs, and as a result, it's been a bit more stable than Vista. That will likely change as time goes on and the bugs in Vista get worked out. Many of them have already been fixed.



"Niche" meaning one of the most ubiquitous electronic devices on the planet, of which gaming is only one small part of its many uses? That's a larger "niche" than any console can ever claim.

I've got devices that are a lot more niche, like my graphing calculator, that have lots of games. Every device I know of that has a general purpose programmable chip has games for it. Even if PCs go totally indie, there is never going to be a shortage of games for the PC. Maybe they won't be quite as "professional," but they're certainly going to continue. If anything, it's the consoles that are a niche: Granted, it's a very large niche, but consoles are nowhere near as general purpose or as popular as PCs. They're definitely more niche than PCs.

Which would you prefer in your game:

1: It leak memory until it crashes after hitting the 2GB limit?

2: It thrashes the garbage collector, quickly rendering the game unplayable until it hits the 512MB limit on Java process memory size and promptly crashes to desktop?

At least with #1, the game lasts longer before it dies. And it can use more than a set limited quantity of memory.



I missed some words there. The free ones I'm talking about are very much high quality. Things like Boost and so forth. Far too few C++ programmers even know about Boost, let alone use it, and that is criminal.



99% of the computers are not part of the computer game industry. The users of those machines will neither purchase nor play PC games. They are not part of the market.

Virtually every console sold is part of the console game industry. Virtually every user of these consoles will purchase and play console games. They clearly are part of the market.

The PC as a device isn't a niche; it's the PC game industry that is a niche market of the larger game industry as a whole.



Once again, "niche" doesn't mean "shortage of games". The two are not equal.

That's nice. There are also commercial libraries that are high quality as well. You missed my point.



And most statistics are made up on the spot.



I seriously doubt it. There is no shortage of PCs with games on them, and no shortage of PC games. Wishing something to be true doesn't make it true. I seriously doubt the PC game market is smaller than the console game market.

I don't know the exact percentage, but the extreme-majority of computers are not used for AA/AAA gaming and-or simple incapable of playing recent AA/AAA games.



What he said before you was kind of true, the PC gaming industry is growing into a niche gaming market every year. Last year, PC Gaming generated less then one-tenth (~7%) of the gaming industry's game sells, and only about 14% of the industry's revenue. Keep in mind, that over one-third of all PC game sells and revenue last year was Blizzard's World of Warcraft and it's expansion The Burning Crusade. And, The Sims 2 took another giant slice of that pie, with over a quarter of all PC game sells.

That means two games and their associated expansion packs account for the majority of all PC game sells. Not to mention there was another decline of the PC gaming industry last year by 6.5% in revenue from 970 million to 910 million. In less then three-years (two for some), each current-gen console has greater game sells and revenue now then PC games. By the end of this year, those console numbers will grow in the double-digits, while PC gaming will drop another 1-10%. Niche enough for you?

I'm not disputing your points, as I've read pretty much the same thing, but if memory serves, those statistics are not including digital sales, only retail sales figures. So it's slightly misleading. Still though, PC gaming is a niche market. But it's a niche I'm comfortable being a part of and when (or if) it dies out, I'll probably stop gaming.

More than slightly, retail might even be the minority share of PC game sales now. Are they even adding Amazon to the stats yet?

Alfonse, do you actually use any consoles or do you just enjoy ragging on PC games? One of the Kotor games flat doesn't work. It gets to a certain spot and a bug prevents further progress. There have been numerous console games with memory leaks that prevented extended play. Numerous games have also shipped with hard crashes that lock up or reboot the systems under the right circumstances. The industry regularly does stories on how shitty various console games are. Prince of Persia:WW is at the top of the list with game ending bugs. Multiple Rainbow 6 games have had severe flaws, the lean glitch is probably the most well known. These aren't minor publishing house titles released by nobodies. For someone in the industry, your ignorance is impressively biased.

Yes, I play on consoles. However, I don't buy crap.



Having played through both KotOR games, I can certainly say that your statement is flat-out wrong. However:

1: Both KotOR games are quite buggy. KotOR II significantly more than KotOR I, but the first was certainly quite buggy.

2: Both KotOR games were made by PC developers, not console specialists. They were built using engines designed and built for PCs.

As more PC developers turn to consoles, it's natural that you will find their habits infecting the console space as well. By contrast, traditional console developers (Capcom, Konami, Nintendo, Namco, Square-Enix) have avoided this.

Haha what? I played through KOTOR 1 and 2 both on PC and Xbox back in the day, and never ever had a problem. If the game was that broken it would have been recalled.

Just takes a google search to find out he's not wrong. Here's a quote:



Granted, that's from way back when the game was first released (from videogamereview.com), but it at least shows psychoak isn't talking out of his butt.