Musings on the state of the beta and project trajectory

[PrimeJunta]

There's been a quite a lot of angst here about the state of the beta, which frankly isn't all that great, what with the disappearing items, disappearing quests, characters freezing in place, and general combat clusterhug thing. I'm a software developer (full-time for 14 years this year, 29 years counting from my first paid project... no, really, I am), and there's a couple of things I think that are worth pointing out.

 

It's clear that the BB was released when it was released due to Gamescom. And as some general put it, you go to Gamescom with the build you have, not the build you want. We can argue all night whether it would've been better for Obsidian to delay the BB until the most critical bugs are sorted or not (and boy would there have been howls had they done that!), but this is the build the showed us.

 

Thing is, unless you're actually working on the project, there is no way to tell anything much about the project trajectory from one data point. You need at a bare minimum two. That means that we'll be a lot wiser when the next build is released. If we get it in a week or so and most of the critical and some of the less critical bugs are fixed and there's not a lot of regression (=new bugs, or re-emerged old bugs), then things are looking positive: the project is on a good trajectory, and any delays or budget overruns that may happen will be minor.

 

If, on the other hand, most of the critical bugs are still there, or they're gone but new equally critical bugs have appeared, then there's cause to worry, because it means they're struggling with the product. I'm not actually a game developer so I can't say anything specific about those bugs, but the general smell I get is that they oughtn't be hard to address. They're in isolated systems, most are easily reproducible, and if the code is at all sane it ought to be pretty straightforward. The pathfinding issues are likely to be the trickiest, I would guess, but the save/load issues and disappearing-item issues ought to be pretty routine. If they're not, it's indicative of architectural or process/practices problems, and those are the real bastards.

 

As an aside, there's also cause for optimism: at least on my fairly standard, middle-of-the-road system, the beta barely ever crashes, freezes, has major display or audio glitches, or similar issues indicative of fundamental problems. I know this is not the case for everybody, but that's almost certainly due to hardware or driver quirks that they haven't had time to test for yet; at this point having some percentage of the systems not be able to run the game at all is to be expected.

 

What I'm sayin' is, chill. For now anyway. Postpone the angst until the next build. It ain't gonna be perfect, but only then we'll have an idea about where the project is headed. If that build is as bug-ridden as this one, or if there just plain won't be one "in a few weeks," I'll join in the hand-wringing. Hey, I've even got a friendly wager running on this, so I've got something extra to hand-wring about.

[sparklecat]

Well said; I, too, have been incredibly pleased with how well my 4 year old system has been handling the game, especially when this is before all the optimisation.  Need to test it out a bit more on my laptop, but even there I was expecting much worse in the way of loading times at this stage.  I'm feeling fairly optimistic as well on this score.

 

Also on the "oh no there isn't enough time for them to fix everything!" score; I think they are likely better placed to judge that than I am right now.

[Eldram]

Well said PrimeJunta, I could not agree more! This is a beta, a simple notion some seem to forget. It should have bugs and a few larger issues, that is kinda the whole point. I would be suprised, knowing their track record, if the finished product is any less then stellar!

[Savant1021]

I have to say that for (what I think is) the first more public build of beta, I feel like what we've been given is rather well-polished, compared to other "betas" I've participated in previously. There are obviously bugs to smoosh--some rather frustrating, to be sure. 

 

I agree that the next build will be a good indicator of progress and handling of the product overall. Fingers crossed that it is very much a net improvement to the stability of the game.

[EdgarAllanBroe]

Not really sure why people are mad about a beta being buggy. By Obsidian no less. Don't get me wrong, I love their games, but their games can be buggy as all hell even by release. I usually let it slide because they do a lot of post launch support, but I still don't know why people are surprised about the beta being buggy.

[Azrael Ultima]

Not really sure why people are mad about a beta being buggy. By Obsidian no less. Don't get me wrong, I love their games, but their games can be buggy as all hell even by release. I usually let it slide because they do a lot of post launch support, but I still don't know why people are surprised about the beta being buggy.

Sometimes it's as if they can't fathom the possibility that an unfinished product could possibly not yet be finished.

 

I think the marketing "geniuses" responsible for using "betas" as a method of generating more publicity are at least partially to blame for that. Too many people got the idea that Betas are really just free demos now.

[Matt516]

 

 

Not really sure why people are mad about a beta being buggy. By Obsidian no less. Don't get me wrong, I love their games, but their games can be buggy as all hell even by release. I usually let it slide because they do a lot of post launch support, but I still don't know why people are surprised about the beta being buggy.

Sometimes it's as if they can't fathom the possibility that an unfinished product could possibly not yet be finished.

 

I think the marketing "geniuses" responsible for using "betas" as a method of generating more publicity are at least partially to blame for that. Too many people got the idea that Betas are really just free demos now.

This. The distortion of the word "Beta" is the result of a lot of the disgruntlement, I think.

[Marceror]

Well said PrimeJunta. I look forward with great anticipation, and hope, for the next release.

[Qiushui]

To me, it's just the bugs that sort of overly inflate the perception that the sky is falling. No one likes bugs.

 

Everything else, like general lack of polish and the need for many, many tweaks to several different areas of the game, is exactly what I expected.

[halethrain]

To me, it's just the bugs that sort of overly inflate the perception that the sky is falling. No one likes bugs.

 

Everything else, like general lack of polish and the need for many, many tweaks to several different areas of the game, is exactly what I expected.

 

A beta is supposed to have bugs. That's the definition of beta: feature complete, and testing for bugs/optimizations. The problem here is that everyone's expectations are not based on the reality of software development. Everyone here seems to believe that they were getting a 5-hour demo, not a beta.

[Azrael Ultima]

To me, it's just the bugs that sort of overly inflate the perception that the sky is falling. No one likes bugs.

So you don't want any of the roasted ants, then?

[Qiushui]

 

To me, it's just the bugs that sort of overly inflate the perception that the sky is falling. No one likes bugs.

So you don't want any of the roasted ants, then?

 

Is that a reference to some quest or item in the beta? Sorry, I've only played about half of the content so far.

[Yonjuro]

There's been a quite a lot of angst here about the state of the beta, which frankly isn't all that great, what with the disappearing items, disappearing quests, characters freezing in place, and general combat clusterhug thing. I'm a software developer (full-time for 14 years this year, 29 years counting from my first paid project... no, really, I am), and there's a couple of things I think that are worth pointing out.

 

 

 

 Nice post. Also, (speaking as another software person), the seeming multitude of bugs could really be a few nasty bugs (some piece of code stepping on some memory that it shouldn't, race conditions in  threads etc.). A single problem can show a very large number of symptoms if it is the right (err, wrong) kind of problem.

[Qiushui]

 

To me, it's just the bugs that sort of overly inflate the perception that the sky is falling. No one likes bugs.

 

Everything else, like general lack of polish and the need for many, many tweaks to several different areas of the game, is exactly what I expected.

 

A beta is supposed to have bugs. That's the definition of beta: feature complete, and testing for bugs/optimizations. The problem here is that everyone's expectations are not based on the reality of software development. Everyone here seems to believe that they were getting a 5-hour demo, not a beta.

 

Nothing of what I said in any way disagrees with a beta having bugs. I'm merely stating that, beta or not, no one likes bugs and it will lead to some amount of complaints, justified or unjustified.

[Azrael Ultima]

Nothing of what I said in any way disagrees with a beta having bugs. I'm merely stating that, beta or not, no one likes bugs and it will lead to some amount of complaints, justified or unjustified.

The thing is that you should have fully expected for there to be bugs. Of course nobody likes them, but if there aren't any, why are they doing a beta in the first place? In a beta, bugs should be reported(if not already well known), not complained about, as that only wastes everybody's time.

[Qiushui]

I did expect for there to be bugs. What I largely didn't expect was bugs that would be borderline game-breaking (with the exception of hardware / driver comparability issues) because that is counter productive to providing actual gameplay feedback, which is also the point of being here.

 

However, I'm not a 5-year old child and I'm fully aware that the bugs are a high priority and will be fixed in due time, as is entirely reasonable.

 

For the record, I have hardly complained about anything in the beta and my impressions have been overwhelmingly positive with what I have seen and the direction the game is going.

[Marceror]

 

To me, it's just the bugs that sort of overly inflate the perception that the sky is falling. No one likes bugs.

 

Everything else, like general lack of polish and the need for many, many tweaks to several different areas of the game, is exactly what I expected.

 

A beta is supposed to have bugs. That's the definition of beta: feature complete, and testing for bugs/optimizations. The problem here is that everyone's expectations are not based on the reality of software development. Everyone here seems to believe that they were getting a 5-hour demo, not a beta.

 

There are bugs, and then there are bugs (i.e. the game breakers). Ideally Obsidian would have addressed the major game breakers during their internal beta process, so that tens of thousands of backers wouldn't have to experience them and report them ad infinitum. You simply don't need 20,000+ people to confirm that gear and quests disappear regularly or that characters often get stuck in place.

 

While I respect PrimeJunta's points, I think most around here realize that I'm firmly in the camp that believes Obsidian jumped the gun releasing this to backers -- and I can appreciate that a software development veteran, who's used to this sort of process sees if differently.

 

Based on Obsidian's own announcements, the backer beta is meant to focus on the classes, combat, UI, crafting/enchanting, etc. But until you have a more functional game it's extremely difficult to provide meaningful feedback on those areas. The fans are not professional beta testers, so giving them a fundamentally broken game and expecting them to know how to sort that out is unrealistic at best.

[Azrael Ultima]

There are bugs, and then there are bugs (i.e. the game breakers). Ideally Obsidian would have addressed the major game breakers during their internal beta process, so that tens of thousands of backers wouldn't have to experience them and report them ad infinitum. You simply don't need 20,000+ people to confirm that gear and quests disappear regularly or that characters often get stuck in place.

Sometimes you do need them to figure out just why the heck it is happening, though. That aside, you need to start at some point. You can't always afford to wait for that "perfect" beta build.

[Marceror]

 

There are bugs, and then there are bugs (i.e. the game breakers). Ideally Obsidian would have addressed the major game breakers during their internal beta process, so that tens of thousands of backers wouldn't have to experience them and report them ad infinitum. You simply don't need 20,000+ people to confirm that gear and quests disappear regularly or that characters often get stuck in place.

Sometimes you do need them to figure out just why the heck it is happening, though. That aside, you need to start at some point. You can't always afford to wait for that "perfect" beta build.

 

Well, we're pretty god-damned far from perfect right now.

Edited August 25, 2014 by Marceror

[Azrael Ultima]

 

Well, we're pretty god-damned far from perfect right now.

 

It's compiling at least. It even starts up, you can get in the actual game and you can play for a while.

[PrimeHydra]

PrimeHunta, let me just say from one Prime* developer to another: You nailed it!

 

Based on what I've seen from responses to the more critical bugs, I think the game will fine. OE were already aware of most of the posted bugs, and I can vouch for a lot of them being trivial (probably).

 

The one and only issue that concerns me at a design level is combat. As another guy put it, PoE feels more pause-with-realtime than realtime-with-pause. This is due mainly to the vast number of commands (i.e., abilities) you need to issue to succeed in combat--and the need to issue them in quick succession.

 

A decent set of AI scripts (think Baldur's Gate where you could choose Ranged, Melee, and a dozen variants) would help. But how would they keep movement in battles quick enough to look natural, without making the player feel they have to frantically hit space lest they lose a tactical opportunity?

 

Well, Diablo's developers did manage to turn a turn-based strategy game into a real-time action game. Maybe Obsidian will do the reverse and build a turn-based option on top of the existing system. I don't know; it will be really interesting to see how this system evolves.

Edited August 26, 2014 by PrimeHydra

[Quietwulf]

Completely agree, although I'm kinda shocked to find a reasonable, well rounded opinion on the internet! 

 

I too share the concerns about the combat system.  It's been an interesting exercise in design, because I think I've come to a conclusion;

 

The more complex the choices asked of the player, the harder it becomes to make those choices in real time.

At the moment, the combat system feels like it has so many moving parts, it can't be played effectively *without* pausing to take in what's just happened and adjust.

 

Compare this to games with real time combat and you find systems that are by in large far more streamlined and easy to understand.

[Guest]

Based on Obsidian's own announcements, the backer beta is meant to focus on the classes, combat, UI, crafting/enchanting, etc. But until you have a more functional game it's extremely difficult to provide meaningful feedback on those areas. The fans are not professional beta testers, so giving them a fundamentally broken game and expecting them to know how to sort that out is unrealistic at best.

 

It is possible (dare I say, likely) that they released what they had knowing that some non-zero percentage of the feedback that they were going to get was going to be garbage.

 

Considering the amount of bitching and moaning that took place before the beta came out, I can't imagine that they are least bit surprised to see bitching and moaning afterwords as well.

 

I guess maybe the only thing to do now is take a long, hard look in the mirror and ask ourselves if we are the people helping or the people bitching.

Edited August 26, 2014 by Achilles

[Sartoris]

I think one of the key class design features that made IE games combat work as well as it did was that your melee/archer characters could effectively be put on cruise control for a little while. They would wail away on the assigned enemy and you could be assured that they were doing an acceptable amount of damage and/or providing appropriate interrupts/penalties for their combatant.

 

This allowed you as the player to focus on making tactical considerations with your casters and other melee classes that required specific positioning e.g rogues.

 

I realize that the designers want all classes to be balanced and all builds to be viable. I don't see why this means that the player themselves must be required to devote the same amount of mental energy to each character. Having five or six characters all requiring as much mental devotion as a normal BG/IWD casting class seems simply exhausting to play and I'm not looking forward to it.

 

Moreover, there are plenty of people that simply wont be as interested in the combat. Giving them the option of running a very melee heavy party, thus reducing the mental tax of combat, would possibly be a very welcome play style for these folks.

 

Finally, if many of the melee classes abilities are turned into toggles or instant fire and forget type buffs then that would present a very different play mode for those classes. Currently, all classes have a wide selection of abilities that cause various positive and negative status effects or damage to the enemy, but the gameplay is similar for multiple classes. Select the character and pick the skill you want to "cast," wait till the character can "cast" again and rinse repeat until all baddies are dead.

 

If melee classes instead were about picking the "combat mode" and self-buffs i.e. berserk, etc that fit the tactical considerations of the fight and then required only a limited mental engagement from the player it would present a very different playstyle than the game's casting classes.

[Mayama]

I dont care how long the first update takes because its the first beta version. They have way more stuff to work on than in the following betas. If the basics work its easy to implement single systems and observe what they will break. I think we will start to get fast updates when the game gets into the balancing phase because its most likley only adjusting numbers in a text file.