What is Beta?

#1 I've struggled with this one myself. Here's what I've come up with based on feedback from my developers and experience:

Proof of Concept - quick and dirty go at basic game play
Alpha - running game without bells and whistles. place holder graphical assets - bugs galour to be squashed
Beta - Most if not all graphics. Full game play. Time to do some End User Testing. There be bugs here. Can have a number of beta interations i.e. 3 or...(place number here) versions of the beta game.
Gold - Final. No (we hope) bugs. All assests in place. Go to the presses.

There seem to be no hard and fast rules that I've been able to find. Perhaps others here can add clear definitions of the process - Please :)

J.

#2 I consider beta to mean:

This is the way the game is going to be when it releases. It's not bug free, at this point, there could be compatibility issues. Could be minor gameplay tweeks, but beta means, to me, pretty done.

#3 Hello,

Pat, has the same understanding as me. Its complete but needs more testing before a full release.

Ben

#4 One of the key points of a beta for online games by the way is that the purpose of "beta" is to do the load/performance testing that you cannot do internally (even with loading tools). Its your first "real" exposure to massive amounts of players, if that applies to your game.

While it doesn't matter for your contract requirements necessarily, there is a huge difference between what developers normally view beta as, and what players do. To developers, a beta is almost feature locked (well, hopefully!), while to players at large, it's the first time they can contribute major feedback on systems designs. This puts the developers in a catch-22: Do you listen to your player's feedback and redesign major systems if required? If so, on your internal definitions, you're now forced to revert back to an alpha, or even Proof of Concept phase (especially if the feedback is huge/very critical). Or, do you ignore player base feedback and criticism, and possibly release a game that won't sell?

Unfortunately, since from the dev's perspective, beta is dead last in the cycle, you don't have time or budget to do this--but the players may very well demand it if systems just aren't working, or don't appeal. Personally, we're trying to figure out a way to have an "open alpha" for a period of time so that this general player feedback happens much earlier in the development cycle, instead of less than a few months before projected gold. This sets up some huge issues of it's own, which is why we are struggling with how to handle it (vaporware claims, too much hype too early, are just two of those issues).

#5 How could I forget "Feature Lock". That is the primary element when going into Beta. Clients (and developers) are prone to "feature creep" and that can't happen once you've reached beta.
But as you say Stephen, if you wait until Beta to do testing you may end up having to make some tough decisions based on feedback.
The problem with showing a client or end user an Alpha (and correct me if I'm way off base here) is having to educate them in understanding "this isn't how it will LOOK when it's done - and yes it's buggy". That can be tough because depending on your testing group, they may colour their opinions with how they feel about the "presentation". I've been through this so many times. You waste a lot of time going over this basic point - "It isn't finished yet".

Sorry, I got off on a ramble there. Very useful information here. I don't know about you Mario but I'm learning something here :)

Cheers

J.

#6 Yes this is exactly what I was looking for.

Thanks