Thursday, January 03, 2008
In Defense of Game Patch
Every once in a while ("a while" being defined as "about five minutes") someone complains about the excessive patches required in modern PC games. I get as infuriated as anyone else, especially for those really nasty cases where the patch invalidates my saved games and makes me start over. Nobody likes seeing their game wrecked by a bug. Why didn't they bother testing before they released the game, anyway? And we're seeing day 1 patches (meaning, patches released at the same time the game hits store shelves) with huge lists of bugs fixed between the time the game went to the duplicators and the time it hits the shelf.
Something has got to be wrong here, right?
And it's no longer just PC games, either. Consoles games, now that they have downloadable content, are beginning to show signs of patch fever. Is this a sign that developers are just lazy, and use the fact that there's now means of distributing a patch as a crutch to release games in a buggier state?
Now, I'm probably an official "part of the problem," because I sympathize with the need for patches. And I'm a developer. And I've released patches for my games. So I'm gonna go on a limb here and actually defend this horrible practice of releasing buggy, broken code.
Well, not really. I can't condone that. But I do want to talk about why your favorite game is on its third patch in nearly as many weeks. It's probably not as bad as it sounds, and the publisher probably didn't just get bored and try to foist off a horrible mess on an unsuspecting public in hopes of getting a quick buck.
Probably. At least not most of the time. I think.
When people talk about how buggy games are, they often compare it to the old days, when you rarely heard about patches or lists of bugs as long as your arm.
Things Haven't Gotten Worse. Well, Not Much
But it wasn't that the games of yesteryear weren't buggy - it's just that there was no reasonable way to distribute a patch. I do remember getting a patch back in 1991 that was "by request only" - they'd fixed the bug post-release, but didn't really advertise the patch, because it cost them money every time they had to send out a floppy disc with the fixed executable. So if you called customer support and asked specifically about a problem addressed in the patch, then they'd send you the updated diskette.
Right, I said diskette. This was back in the day when the entire code and data could fit on a couple of 1.4 meg floppies.
Nowadays, many games are about three orders of magnitude larger than that. And we haven't suffered through an increase in bugs of similar magnitude, which says a good thing. But you can't compare Pong to Oblivion and say, "Oh, look, they are so sloppy these days... look at how many bugs Oblivion has compared to Pong."
I mean - it's Pong. You can test the whole game in about five minutes. And you STILL had bugs in some games in that era with scores that would wrap around from 255 to 0. Or that bug in the arcade game "Sinistar" (yes, in the arcade) where a player could - on demand - get himself shot simultaneously with getting himself eaten by Sinistar when down to only one ship... which resulted in 2 kills at the same time, taking his number of lives UP to 255. I don't know if they ever patched that one. It still has that bug in an emulated version of the game (complete with a video where one of the developers talks about the bug).
In the past, they simply tried to keep the issues secret to keep their happy customers from knowing that there were problems. Nowadays, there's more open communication, so the happy customers discover that all might not be well in their games. Personally, I'm for better communication, even if it might alarm customers.Hey, I've been on the disaffected customer side of things, too. It happened to me with no less than Ultima VII part 2: Serpent Isle. To this day I have never completed it. I'll bet most of you old-timers who played Serpent Isle had no idea there was a game-killing bug that only seemed to affect players running the game on a Cyrix 386/40 CPU (according to the tech support guy who could only offer apologies to me at the time).
What Gets Fixed?
When you look at the list of bug-fixes to the game in the latest patch, if you are being completely honest, how many of those bugs were you actually aware of in advance if you don't frequent the message board forums to hear people grouse about them? If you were playing the game in a vacuum, oftentimes you may be completely oblivious to every single bug in the list. There may be some issues that you only recognize after-the-fact as being bugs that you may have only chalked up to some design quirk ("Oh, so that's why those healing potions didn't seem to work so well...").
Then there's the gameplay tweaks that get updated with patches. Little irritations in the game that weren't really bugs, but things that the larger player base recognized as being a detriment to fun. While I wish that all developers and QA people could see the forest for the trees all the time when developing and testing a game, the truth is that more eyes are always going to see more things. When you are working with an evolving game for a long time, you get used to things that you don't realize that you are being forced to get used to. Some recent indie game examples include the arrow patch for Eschalon: Book 1 (to increase the quantity of arrows available in shops), or the rogue enhancements in a recent patch to Depths of Peril.
Then there's the brand-new gameplay features and major enhancements that get thrown in. Things to make a (hopefully) good game even better. You might complain, "why didn't they do that in the first place?" If it's a good idea now, it was a good idea before the release, right?
Well, that brings us to another subject. Why modify a game after it's been released, anyway? They didn't do it back in the pre-Web days, so why should we do it now?
Motives For Patches
First of all, developers and publishers don't want people to be unhappy with their game. Back in the pre-web days, there wasn't much that could be done about it if you found out that 1% of your customers couldn't finish the game due to a game-killing bug. But today, patches provide developers with the means of solving that problem, at least for those customers who are online and pay attention to such things. It makes developers happy (usually) to solve problems, and it makes the customers happy, so it's a win.
Then there's also the very sad truth that games are never, ever finished. At least not these days. They are simply released. If the developer had to make sure it was perfect, the game would never be released. That's reality. Considering the number of "director's cuts" and "special editions" of movie DVDs that are out there, I suspect it's not unique to games. If a game goes out in it's "good enough" state, but then an opportunity (and motivation) comes along to make it that much better, it is another win / win scenario for both developers and players.
Then there's the fact that the players know what they want better than the developers do. When the players identify some "low-hanging fruit" improvements that the developers were unaware of, this becomes a great chance to make a game better with little effort.
Finally, there's a significant ulterior motive for releasing patches. Quite simply, it's a way to keep a game current and newsworthy. Public attention in the gaming hobby is both fleeting and fickle. As soon as a game become yesterday's news, its sales begin dry up. The release of a patch gives everyone an excuse to mention the game again, which in turn may rekindle excitement about it and - hopefully - generate more sales. Or maybe just remind people of how great the game was to keep them interested in an upcoming expansion or sequel. Patches may be more about marketing than anything else.
So if you get bitten by a game-interfering bug, feel free to grouse. But don't take the prevalence of patches as an indicator that code quality has gone to pot across the industry. I mean, okay, I've written some of that code, and I know how bad it really is, but I really do believe modern practices has made it better rather than worse.
(Vaguely) related bugliness:
* On Game Engines and Swarm Missiles
* My Worst Bug Ever
* Twisted Metal Trivia
* Why Software Design Isn't Like Architecture
.
Labels: programming
- posted by The Rampant Coyote @ 6:02 AM (Permalink)
Comments:
Links to this post:
<< Home
I am ambivalent about patches to fix bugs. I think that it's better to have them than not have them, with the caveats that they are NOT a stand-in for extensive beta-testing, and that they should NEVER cause me to lose my saved games. (In the rare case where the latter is not possible, a saved-game converter should be made available) I do think that day-1 patches are an abomination; a real sign of poor testing procedures.
Patches to alter game play trouble me. Yes, games are never finished, but neither are novels, movies, poems, etc. There will always be strong points and weak points. It's important to leave a game the way it was when it shipped rather than worry constantly whether it was balanced right. Too few arrows in the shops? Instead of fixing it, let it be part of the character of the game. It's not "broken", it's just harder to play an archer character. This is the same fanbase who thought it was fun to play through Final Fantasy I with a party of all white mages, after all.
Alternately, (and this works better for indie developers) you can have a beta+1 cycle, where you basically sell a beta for a month or two, with the understanding that there will be a more intensive patch cycle for those months involving lots of tweaks and changes, and culminating with a final release after which only bug fixes will be patched.
Patches to alter game play trouble me. Yes, games are never finished, but neither are novels, movies, poems, etc. There will always be strong points and weak points. It's important to leave a game the way it was when it shipped rather than worry constantly whether it was balanced right. Too few arrows in the shops? Instead of fixing it, let it be part of the character of the game. It's not "broken", it's just harder to play an archer character. This is the same fanbase who thought it was fun to play through Final Fantasy I with a party of all white mages, after all.
Alternately, (and this works better for indie developers) you can have a beta+1 cycle, where you basically sell a beta for a month or two, with the understanding that there will be a more intensive patch cycle for those months involving lots of tweaks and changes, and culminating with a final release after which only bug fixes will be patched.
Well, I know for a fact that some publishers push a game out in a buggy state sometimes, especially to hit the Christmas season. A game released in late November with a lot of bugs is very, very suspicious. I think in those cases, yeah - the publisher is banking on patches to fix a bad release. And that's heinous. If you use patch distribution as an excuse to turn your paying customers into paid QA people, that's wrong.
But the best QA team in the world isn't going to be as effective as the eyes of thousands (or hundreds of thousands) of players running on a nearly equal number of hardware variations. Again - we have to deal with reality.
The altering game play thing doesn't usually bug me, but I can see the point for some types of games. If you're playing an RPG and you have optimized your character around a particular set of strategies that are suddenly rendered ineffective, that's a problem. But I rarely see changes that are that extensive in single-player games. MMO's... well, I don't see them there, either, though players like to blow such tweaks way out of proportion.
But the best QA team in the world isn't going to be as effective as the eyes of thousands (or hundreds of thousands) of players running on a nearly equal number of hardware variations. Again - we have to deal with reality.
The altering game play thing doesn't usually bug me, but I can see the point for some types of games. If you're playing an RPG and you have optimized your character around a particular set of strategies that are suddenly rendered ineffective, that's a problem. But I rarely see changes that are that extensive in single-player games. MMO's... well, I don't see them there, either, though players like to blow such tweaks way out of proportion.
Well, I agree with the "gamer-friendly" comments you made but a lot of reasons why some major games suck isn't very relevant to the buyer - the one with the money, right?
However, when looking at major "fuck-ups" one can surmise that RPGs are a prime example what can go wrong - simply because of the long developemt time and the possibilities of problems with the companies behind them.
Here are some example from the top of my head:
Ultima 7 and 7b (Serpents Isle) - the former having most problems with its memory manager and the latter being rushed to market even though most technical problems had been eliminated.
Ultima 8 needed a big patch that changed the type of gameplay (no "Super Avatar Brothers" with timed jumps anymore) but at least the game was finishable in its original form.
Ultima 9 had an official "final patch" after EA abandoned its support. Now the community tries to reprogram the whole mess without the bugs.
Anachronox had several patches long after Ion Storm abandoned the game - they were made by a single developer (AFAIK) of the game who had a conscience.
Vampire the Masquerade - Bloodlines:
Official patches until version 1.2 dated Dec. 2004, then Troika folded(?) and community patches until this day - version 4.4 dated Dec. 2007 - and still not everything fixed (AFAIK).
Gothic 3, the German Gerstmangate:
Rushed to market for christmas 2006, many packages sold and mixed reviews - because it was too damn buggy.
The development team Piranha Bytes abanoned the ship after it became known what practices publisher JoWood used with the press and how it tried to make Piranha Bytes accountable for the mess.
The result was no more patches by the developer and community patches for a year since JoWood wasn't "able" to help its customers.
In spite of this whole trouble (and many boycots of potential customers) and probably because of the controversy the game sold very well so the industry will surely continue on its path to less quality.
And, yes, Oblivion - playable without a patch (except a few side quests) but the German localization sucked big time and the community had to patch that.
You also mention game consoles - well, I lost my interest in them as soon as they had "downloadable content". Not that there weren't buggy games on older platforms before but usually the games were QA'ed enough to be acceptable.
Take care,
Calibrator
However, when looking at major "fuck-ups" one can surmise that RPGs are a prime example what can go wrong - simply because of the long developemt time and the possibilities of problems with the companies behind them.
Here are some example from the top of my head:
Ultima 7 and 7b (Serpents Isle) - the former having most problems with its memory manager and the latter being rushed to market even though most technical problems had been eliminated.
Ultima 8 needed a big patch that changed the type of gameplay (no "Super Avatar Brothers" with timed jumps anymore) but at least the game was finishable in its original form.
Ultima 9 had an official "final patch" after EA abandoned its support. Now the community tries to reprogram the whole mess without the bugs.
Anachronox had several patches long after Ion Storm abandoned the game - they were made by a single developer (AFAIK) of the game who had a conscience.
Vampire the Masquerade - Bloodlines:
Official patches until version 1.2 dated Dec. 2004, then Troika folded(?) and community patches until this day - version 4.4 dated Dec. 2007 - and still not everything fixed (AFAIK).
Gothic 3, the German Gerstmangate:
Rushed to market for christmas 2006, many packages sold and mixed reviews - because it was too damn buggy.
The development team Piranha Bytes abanoned the ship after it became known what practices publisher JoWood used with the press and how it tried to make Piranha Bytes accountable for the mess.
The result was no more patches by the developer and community patches for a year since JoWood wasn't "able" to help its customers.
In spite of this whole trouble (and many boycots of potential customers) and probably because of the controversy the game sold very well so the industry will surely continue on its path to less quality.
And, yes, Oblivion - playable without a patch (except a few side quests) but the German localization sucked big time and the community had to patch that.
You also mention game consoles - well, I lost my interest in them as soon as they had "downloadable content". Not that there weren't buggy games on older platforms before but usually the games were QA'ed enough to be acceptable.
Take care,
Calibrator
# posted by 
: January 03, 2008 3:06 PM
: January 03, 2008 3:06 PM
Hey, I know how to crash Twisted Metal on the Playstation. The QA guys caught the bug on the PC, but it had already shipped for the Playstation. I don't know if anybody ever managed to find that crash out in the wild, though.
For what it's worth, I really only encountered one major bug (resulting in an impossible-to-complete subquest) in Vampire the Masquerade: Bloodlines. I did play it patched with the last "official" patch, though.
So that's another one where people rip on how buggy it was, but I wasn't really affected by that.
Now... you wanna talk buggy releases... Falcon 3.0, released in 1991. They DID release a series of patches for it, back in the day where you had to download the patches off of a BBS at 5600 baud or something like that. But that one - as released - had uncompletable missions, tanks that drove on the ocean, MiG-19s that were super-planes, and you were assigned to protect units that were already dead.
For what it's worth, I really only encountered one major bug (resulting in an impossible-to-complete subquest) in Vampire the Masquerade: Bloodlines. I did play it patched with the last "official" patch, though.
So that's another one where people rip on how buggy it was, but I wasn't really affected by that.
Now... you wanna talk buggy releases... Falcon 3.0, released in 1991. They DID release a series of patches for it, back in the day where you had to download the patches off of a BBS at 5600 baud or something like that. But that one - as released - had uncompletable missions, tanks that drove on the ocean, MiG-19s that were super-planes, and you were assigned to protect units that were already dead.
I'm with you -- Patches posted the day after a game is released just tells me that existing bugs weren't fixed because the scheduled release date was met.
This is really bad.
And often, I've seen patches approaching 300 mb. 300 mb! In my opinion a game should never be released if so much needed to be fixed.. it's an inconvenience and an insult to the customer, and nothing more than a "we'll take your money now and maybe fix the game later, if you're lucky and we find the time" situation.
I don't mind patches that alter the game play in a reasonable way, but if they fundamentally change the game then there is a problem.
This is really bad.
And often, I've seen patches approaching 300 mb. 300 mb! In my opinion a game should never be released if so much needed to be fixed.. it's an inconvenience and an insult to the customer, and nothing more than a "we'll take your money now and maybe fix the game later, if you're lucky and we find the time" situation.
I don't mind patches that alter the game play in a reasonable way, but if they fundamentally change the game then there is a problem.
# posted by : January 04, 2008 8:17 AM
: January 04, 2008 8:17 AM
300 MB patches (Gothic 3 had a rumored patch of 1,5 gigs...) do show one thing: The data structures and program logic was not finalized.
Mostly patches are a synonym for "bugs" or programming mistakes but there is an incredible amount of content (be it 3D models, textures, artwork...) that should have been finalized long before the whole caboodle went gold.
Accessing the data wrongly is one thing, delivering loads of wrong data is a simple proof for unfinished work.
Take care,
Calibrator
Mostly patches are a synonym for "bugs" or programming mistakes but there is an incredible amount of content (be it 3D models, textures, artwork...) that should have been finalized long before the whole caboodle went gold.
Accessing the data wrongly is one thing, delivering loads of wrong data is a simple proof for unfinished work.
Take care,
Calibrator
# posted by : January 04, 2008 9:13 AM
: January 04, 2008 9:13 AM
Gigantic patches like that are just insane. At least for a bug-fixing patch. If it's like a whole friggin' new expansion or something they are giving away for free, that's one thing. But this is something else entirely.
I have also heard rumor of a publisher screwing up and delivering the wrong disc to the duplicators. I don't remember what game it was, but that sounded like a screw-up of pretty major proportions if it was true. I think the right thing to do in that case would have been to issue a recall on the discs and replace them with the correct versions.
Which mighta tanked the publisher, so what's right might not have been what was realistic, but still.
Besides Ultima 7 part 2, two of my biggest bug-based disappointments were Frontier: First Encounters (kinda-sorta "Elite 3") which was pretty much unplayable with a joystick, and "Twilight:2000," by Paradigm and published by Microprose, which was an okay RPG up until the climax, where it crashed on every machine I tried it on.
I have also heard rumor of a publisher screwing up and delivering the wrong disc to the duplicators. I don't remember what game it was, but that sounded like a screw-up of pretty major proportions if it was true. I think the right thing to do in that case would have been to issue a recall on the discs and replace them with the correct versions.
Which mighta tanked the publisher, so what's right might not have been what was realistic, but still.
Besides Ultima 7 part 2, two of my biggest bug-based disappointments were Frontier: First Encounters (kinda-sorta "Elite 3") which was pretty much unplayable with a joystick, and "Twilight:2000," by Paradigm and published by Microprose, which was an okay RPG up until the climax, where it crashed on every machine I tried it on.
Come to think of it, I think the publisher DID send a replacement disc for Frontier: First Encounters. But I don't remember the updated version fixing my problem.
Some people say Frontier got fixed, others say it never was.
However, I didn't play it as I never was a big Elite fan (apart from the technical side).
Wasn't 'Temple of Elemental Evil' also a bugfest? I recently bought a budget version but haven't had time to play it yet.
Take care,
Calibrator
However, I didn't play it as I never was a big Elite fan (apart from the technical side).
Wasn't 'Temple of Elemental Evil' also a bugfest? I recently bought a budget version but haven't had time to play it yet.
Take care,
Calibrator
# posted by : January 05, 2008 12:24 PM
: January 05, 2008 12:24 PM
I have played a quite good game with a one day patch. It was actually started after a whole slew of bugs were found in the beta. by the end of the beta, the standard response to balance and bug complaints was "we fixed that in the patch, but it's still being approved by Microsoft. unfortunately, after the patch was relesed it turned out that it actually wrecked game balance once people knew how to counter certain strategies. Patch 1.2 is in progress.
# posted by : January 06, 2008 11:00 AM
: January 06, 2008 11:00 AM
I also recently bought the Temple of Elemental Evil - which is becoming increasingly hard to find (but I learned my lesson from Wizardry 8 - though I don't expect it'll ever be in THAT short supply). I haven't actually INSTALLED it yet and tried it out. Too many indie games (and Gal Civ 2) to dominate my limited play time the last three months.
@(name here) - Yeah, there are sometimes some really weird things that go on with the lead time for games released through traditional channels. From the time a game in for approval to the time it appears on store shelves, several weeks can pass.
The dev team might NOT have just been sitting on their hands the whole time, especially if they are working on additional versions / languages of the game. Or the next expansion... Bugs WILL be found.
As one of my computer science professors once said, "The only bug-free code is code that is no longer being run by anybody."
Post a Comment
@(name here) - Yeah, there are sometimes some really weird things that go on with the lead time for games released through traditional channels. From the time a game in for approval to the time it appears on store shelves, several weeks can pass.
The dev team might NOT have just been sitting on their hands the whole time, especially if they are working on additional versions / languages of the game. Or the next expansion... Bugs WILL be found.
As one of my computer science professors once said, "The only bug-free code is code that is no longer being run by anybody."
Links to this post:
<< Home
