IT'S ALPHA! (Game dev cycles)

[Chroma 61]

ITS ALPHA!

This one phrase gets bandied about with disturbing regularity on these forums, and oftentimes its towards a user who perhaps doesn't know what this really means.
I mean its fairly obviously an Alpha, it says so when you start the game, users on forums keep screaming it at you whenever you bring something up. But what exactly is an Alpha release?

Software development goes in cycles.

1. Conceptual design
2. Alpha candidate
3. Beta candidate
4. Release candidate

You start by drafting up a conceptual design.
In this case, i want to build a platform game, where you can destroy blocks, gain powerups, kill enemies and your objective is to make it to one side of the map to the other.

In college or university they then tell you to move on to drafting up a solid design document, make some nice little flow charts, knock up some psudocode and the like.
This of course rarely occurs past the initial design document (if even that gets done) outside of industry studios people more often than not just move on to knocking out code and assets.
Mostly enough to get you to your first working Alpha candidate.

So you move on from having your idea to fleshing it out into something looking like this: THIS IS AAAALLLLLLLPHAAAAAAAA!!!!!

 

Where light blue is your non-interactive background, medium blue is your player, dark blue a platform, black is the ground and pink is simply another platform. (colors don't really matter, they're merely placeholders)
You move from there to working on your physics, how the player moves around, jump heights, fall rates and basic collision detection.
Once that's all mostly ironed out you can move on to adding more features, like destructible platforms, boxes for power-ups and triggers etc.
Winding up with something like this:

Where dark blue is now destructible, light green is a trigger box and dark green is a box that's been triggered. Graphics still suck because, who cares? It's alpha! Simply to test your design and allow for rapid changes, sure the jump mechanic is a little off, but that's just some bug from loop timing, the controller occasionally seizing up is another, as is the wrong beep playing when you hit a box that shouldn't trigger any audio, along with the juddery frame rate. You can fix those later in your Beta candidate, cause in alpha phase all you're focusing on is BULKING FEATURES, the only bugs you need be concerned with are game breakers that prevent you from testing more features, like not being able to jump at all, or collision glitches that cause you to fall through platforms and the like. Pick a feature, focus on it, get it added to a rudimentary level and rapidly move on to the next, raw ninja coding filled with dirty painful to read through hacks, but it kinda gets the job done!
Fine tuning comes AFTER ALPHA, so you can focus entirely on things like adding power-ups till you get to this stage:

Where you've added your power-ups, got some rudimentary animation like the yellow power-up box being able to move and be collected, all the features you wanted are now added, so your officially into BETA where now you just need to fine tune things, work on the sketchy jump mechanic bug, fix that weird glitch that seems to effect every 17th press of the jump button, and figure out whats causing the controller to occasionally lock up and essentially do nothing.
This is the stage that's most boring, all you'll be doing is testing and bug fixing, reading through code, finding more eloquent and efficient ways of getting thigs to work not just properly but efficiently!

Simples! once all that's worked out, its time to really go about polishing things, cause it looks boring, time to start skinning textures to your collision boxes and make it look pretty, remove the ear destroying beeps that trigger when you collide with certain boxes to nice audio samples and then you'll have your Release candidate.



So that's pretty much it, Design is never really finalized Even at RC stage, you can still punt some DLC ;) you start off with a rough idea and refine and add/remove/alter things as you develop, this should all be done in the broke-assed alpha stage, but often isn't (so be prepared for more game breakers in your Beta phase ;) its real common)
And the Mario example posted is very oversimplified, add in threedee geometry and netcode and things are an order of magnitude more complicated. The premise remains the same however.

So now you should have some idea of what will and wont be fixed during an Alpha cycle, unless its a major bug, it'll be ignored till the Beta stage. And that's why when you post about something seemingly arbitrary, people will cry alpha at you.
Don't take it personally, because regardless of in game kos, banditry antics, the people on the forums are far more civilized and nice. (Everyone's allowed to be unrealistically optimistic occasionally, right?)


 

[frosti 2165]

THEY WILL PATCH IT

[Element47 2480]

you make far too much sense for this forum any internet forum. have beans and hugs.

[Gdaddy22 299]

Also Alpha is so that developers can improve the game and fix things. It's annyoing when people point out the game's flaws or what should be changed and people respond HURR DURR IT'S ALPHA GRAVEYARD PLS.  The people creating those threads are doing good job. They remind us and developers what is there to change. If no one creates any threads about what sucks in the game then the developers can't fix it because they don't always see it from a gamer's perspective and the game could suck indeed. Before you yell IT'S ALPHA learn the difference between stupid threads and threads that point out the game's flaws or weak points 

Edited June 27, 2014 by Gdaddy22

[scriptfactory 620]

You realize that DayZ doesn't even meet your criteria for having a proper development cycle, right? I, honestly, do not believe DayZ has a "solid design document" due to the way development has progressed and based on Rocket's comments during livestreams. He says things like, "Well... we are thinking about adding this" or "I don't know if we will add that. It might be better if the community takes care of it." There is also the small matter of things like KoS, server hopping, ghosting, etc. All of these topics would have been contained in a design doc.

Also, they keep fixing both major and minor bugs in the current (alpha) stage.

So, in my opinion, I think you are wrong with your assessment of a game development cycle as it applies to DayZ.

[Chaingunfighter 917]

You realize that DayZ doesn't even meet your criteria for having a proper development cycle, right? I, honestly, do not believe DayZ has a "solid design document" due to the way development has progressed and based on Rocket's comments during livestreams. He says things like, "Well... we are thinking about adding this" or "I don't know if we will add that. It might be better if the community takes care of it." There is also the small matter of things like KoS, server hopping, ghosting, etc. All of these topics would have been contained in a design doc.

Also, they keep fixing both major and minor bugs in the current (alpha) stage.

So, in my opinion, I think you are wrong with your assessment of a game development cycle as it applies to DayZ.

It's a generalized assessment, it doesn't mean that the devs aren't allowed to fix smaller bugs when they get the chance to. Killing on sight isn't exploiting or an issue with the game, it's a "problem" (not really) with the mentality of players, so you can't really make a plan for dealing with something that's always going to be there. Server hopping & ghosting are exploits, but you're assuming that these side effects were anticipated right from the get go. Remember that the mod didn't have a lot of public hive servers around, so hopping to another server to get loot would put you right back at square one, most of the time. That, and loot was much more abundant and less in depth in the mod, so it wasn't really necessary.

Ghosting is another issue altogether, but they've brought forth some ways of trying to solve it. It's not perfect, but they've done a good job of making sure that people who try to ghost or combat log can't do so with absolute simplicity but people who are legitimately logging in or out aren't punished.

You've also got to understand that rocket isn't exactly the "over-confident" kind of developer that you tend to see. He's an awesome guy and he's got awesome ideas, but he doesn't try to put them forth in the same ways that other people do. When relating to features, you've also got to remember that they want to support modding in the long run, so they have to make decisions on what should be a base game feature and what could probably wait until someone else makes it themselves.

Progress has been good in the 6 month span the standalone has been out.

[xalienax 621]

Also Alpha is so that developers can improve the game and fix things. It's annyoing when people point out the game's flaws or what should be changed and people respond HURR DURR IT'S ALPHA GRAVEYARD PLS.  The people creating those threads are doing good job. They remind us and developers what is there to change. If no one creates any threads about what sucks in the game then the developers can't fix it because they don't always see it from a gamer's perspective and the game could suck indeed. Before you yell IT'S ALPHA learn the difference between stupid threads and threads that point out the game's flaws or weak points 

this a million times over. Many of the topics/complaints that get "ALPHA" responses are not about missing/incomplete features but DESIGN DECISIONS that the poster (and often many others) would not want to see make it to final release.

[scriptfactory 620]

I don't completely disagree with your little write-up but I can tell you that DayZ does not follow a typical software development cycle at all and I doubt there was an actual game design document written up when development started.
 

Killing on sight isn't exploiting or an issue with the game, it's a "problem" (not really) with the mentality of players, so you can't really make a plan for dealing with something that's always going to be there. Server hopping & ghosting are exploits, but you're assuming that these side effects were anticipated right from the get go. Remember that the mod didn't have a lot of public hive servers around, so hopping to another server to get loot would put you right back at square one, most of the time. That, and loot was much more abundant and less in depth in the mod, so it wasn't really necessary.

The griefing and KoSing were out of control in mod. I quit playing because of it and the hacking. In several livestreams Rocket touches on the issue and gives his ideas on how he wanted to control it. He also added several mechanisms aimed at reducing KoS.
 

You've also got to understand that rocket isn't exactly the "over-confident" kind of developer that you tend to see. He's an awesome guy and he's got awesome ideas, but he doesn't try to put them forth in the same ways that other people do. When relating to features, you've also got to remember that they want to support modding in the long run, so they have to make decisions on what should be a base game feature and what could probably wait until someone else makes it themselves.
Progress has been good in the 6 month span the standalone has been out.

I, personally, don't think progress has been good. In the last 6 months we have seen the DayZ dev team work more on the game engine and content (e.g. hats) than game mechanics. For software that has reached alpha status you generally want a complete vertical slice of all features that will be in the public release. DayZ is missing a lot of features that are planned for the beta release.

I am not saying I do not enjoy DayZ. But as a software developer I feel they are developing this game in a very amateurish fashion.

[Chaingunfighter 917]

I don't completely disagree with your little write-up but I can tell you that DayZ does not follow a typical software development cycle at all and I doubt there was an actual game design document written up when development started.

 

The griefing and KoSing were out of control in mod. I quit playing because of it and the hacking. In several livestreams Rocket touches on the issue and gives his ideas on how he wanted to control it. He also added several mechanisms aimed at reducing KoS.

 

I, personally, don't think progress has been good. In the last 6 months we have seen the DayZ dev team work more on the game engine and content (e.g. hats) than game mechanics. For software that has reached alpha status you generally want a complete vertical slice of all features that will be in the public release. DayZ is missing a lot of features that are planned for the beta release.

I am not saying I do not enjoy DayZ. But as a software developer I feel they are developing this game in a very amateurish fashion.

It's definitely unconventional, but Bohemia's always worked that way, to an extent. The reasoning behind engine focus is so they can allow a greater amount of and depth to mechanics. I mean, they really didn't need to make the standalone on a new engine if they were just going to add new features. Yeah, they've been focusing on building the engine up, but that's so many of the future elements work well. Working physics and ragdoll simulation, pathfinding, 64bit servers, and all of the optimization they've been doing are necessary for enhanced systems and newer things like vehicles or AI.

Again, it's not the conventional model, but to deny the progress they've made because they didn't do it as it's usually done isn't a fair statement.

 

Keep in mind that the "play the game in alpha stages" is also a relatively new model of gaming, so you're getting feedback during the major development stages in addition to the game release. It's going to heavily influence your development schedule. That's why we're getting things like Cowboy Hats or MP5Ks in between updates, because they already have the full staff at work.

I'm not saying I support this kind of development layout, but they've made fairly good progress with it thus far, even if it is in engine achievements rather than mechanic achievements, so it's not worth worrying about.

[Chroma 61]

I wholeheartedly agree about the lack of design document, if a full fledged design was in place then there would have been a far less painful transition with the influx of bodies, because everyone would in effect know their role and what was needed and required.
This is why large studios insist on it, because you can replace staff rapidly or outsource parts if needed in order to meet target deadlines.
As a coder in a large studio you're nothing more than a mindless drone, banging out functions to meet defined specifications. The tradeoff is a lack of real creativity or ability to really deviate from the spec.

DayZ is very fluid in its design process, this can be both a blessing and a curse. Open alphas typically require that degree of flexibility to facilitate change and growth in keeping with communal demands, but without a concrete plan it slows down the development, simply because people need to be constantly updated on their function within an organisation as decisions shift to match the current landscape. There's more room for interpretation and confusion.
I'm not faulting the dev team in any way here, or attempting to detract from what they're doing, nor even suggesting a better way of operating, im aware its unusual (and more indie) and I'm entirely fine with that.
They have my beans :)

I was simply trying to help others interpret why their thread got jumped on by "ITS AAAAALLLLPHAAAAA!!!" because the forum is littered with those kind of posts and as a new player who perhaps doesn't understand, it can be fairly intimidating.
I have friends for instance who flat out refuse to post here, simply because of the alpha bandwagon.
They do however lurk.
I just feel that's a shame because for every silly idea there could be a gem that never had the confidence to write a post, The community loses out in such cases. To the detriment of everyone.

Edited June 27, 2014 by Chroma

[hombrecz 832]

I have to say I'm irked about mods closing "the correct caliber" thread. Don't understand why?

 

Also recent post from Ch.Torchia  on reddit made me feel tad uneasy since it clearly shows devs are not changing their approach to "caliber streamlining", hence alpha argument is not helping much.

http://www.reddit.com/r/dayz/comments/298e21/will_we_see_the_addition_of_new_cartridges_or/ciijcw7?context=3

 

p.s. I guess this post is way offtopic hm?

[Nocturnal (DayZ) 85]

People who are upset about lack of progress/bugs at this stage just shouldn't have bought into the game at this stage. Everyone was actively discouraged from buying the game by the developers and no one can blame anyone but themselves if they are upset with the direction and overall pace of development.

 

Bugs/issues shouldn't be ignored, however. But the correct (only) place to properly report these issues is the feedback tracker.

 

The daily rants on this forum and others are just futile - albeit entertaining.

[scriptfactory 620]

Everyone was actively discouraged from buying the game by the developers and no one can blame anyone but themselves if they are upset with the direction and overall pace of development.

 

How many people read the warnings before purchasing the game? I convinced a bunch of my friends to buy the game after putting in hundreds of hours and none of them even lasted for 24 in-game hours. They totally didn't read the warnings, they just listened to what I and the rest of the internet was saying.

Hell, I think Rocket saying "don't buy this game" actually MADE some people buy it.

Edited June 27, 2014 by scriptfactory