Our Properties: Gamasutra GameCareerGuide IndieGames Indie Royale GDC IGF Game Developer Magazine GAO
My Message close
Latest News
spacer View All spacer
 
February 10, 2012
 
DICE 2012: Activision's Hirshberg believes creative people should lead companies
 
DICE 2012: EA's Galda says television's episodic model is the future of game narrative
 
GDC 2012 reveals Super Mario 3D Land, Resident Evil Revelations postmortems
spacer
Latest Features
spacer View All spacer
 
February 10, 2012
 
arrow Virtual Goods - An Excerpt from Social Game Design: Monetization Methods and Mechanics [1]
 
arrow Principles of an Indie Game Bottom Feeder [21]
 
arrow Postmortem: CyberConnect 2's Solatorobo: Red the Hunter [1]
spacer
Latest Jobs
spacer View All     Post a Job     RSS spacer
 
February 10, 2012
 
Retro Studios
RETRO CONTRACT - Environmental Artist
 
Retro Studios
RETRO - CONTRACT AI Engineer
 
Adhesive Games
UI Technical Artist
 
Adhesive Games
Technical Artist
 
Adhesive Games
Senior Network Engineer
 
Adhesive Games
Senior Engine Programmer
spacer
Blogs

  Improving Gameplay Testing In The Console Game Industry
by Travis Johnston on 08/07/09 05:42:00 am   Expert Blogs   Featured Blogs
7 comments Share on Twitter Share on Facebook RSS
 
 
  Posted 08/07/09 05:42:00 am
 

Seems like the average console game company's gameplay testing process starts off with an email like this...

We’re having a playtest this Saturday afternoon, from 1:00 – 4:00. If any of you have gamer friends that would like to participate, have them fill out this NDA + survey and send it back to me.

People showing up are sat down in from of a dev box with the latest copy of the game and promised free pizza if they finish the game and fill out all the questionnaire.  Sometimes if the company is really on it, the testers play is recorded so it can be double checked and reviewed later by all the developers.  This testing is usually < 100 players and is typically only done only 3-4 times during the project at milestones like first playable, alpha, beta.   From this very small sampling of data feedback is given to the development team and at many publishers it is even used to help estimate pre-order numbers and determine a games slice of the marketing budget.

Its not that they do not get any useful data out of this at all, but nothing pulled out of this could be called qualitative and if someone tried to pass this off for testing in any other industry, they would likely be fired.

It is not the developers or publishers fault that there is little to no real focus testing, they do not really have a choice.  Console manufacturers only allow pre-release game disks to only run on special $10k development boxes that the public is not allowed to have.  So even though there are millions of xBox360's out there, your testing pool is still only as many development boxes your studio plus publisher can allocate to it.

Even if your publisher has a ton of money to put in hardware, it is such a issue to coordinate getting that many people on site it can still only be done once in awhile.  The results of the game not being tested that often is so many changes are getting tested all at once that it is hard to determine or isolate what change is effecting the feedback.  There is also problems that we can not verify what live testers say about their playing habits or any other information they give us.

What has this done is placed undue burden on the game designers as they have to be able to come up with new and exciting game designs and get it exactly right as they will get very little chances for any real significant unbiased feedback before they ship.  With the punishment for unpopular levels being a pink slip is it any wonder that designers tend to go with tried and true designs that they are sure will work.  They might be more willing to take risks and move games foward if they had a quick feedback loop to tell them if they are heading in the right direction.

We need the kind of testing online PC games can pull off:

  • Easy distribution of game .
  • Connection to running games so test info can be put in database.
  • Stats on user to ensure they fit our testing needs.
  • Online way to fill out survey and give feedback

We could have this kind of testing because the console manufacturers already have everything we need, we just do not have access to it.  They need to let the developers get access to the play stats on xbox Live so we can pick the right testers for our games.  Then they need to allow us to release xBox arcade type pre-releases of the games so users can play test them for us.

It could work like this:

  • You create you're xBox Live account and you select if you would be interested in getting access to games early for gameplay testing.  It warns them that their stats could then be searched by developers.
  • If you say yes you get put in a database and your playing stats of games is recorded so developers can decide if your good for the type of testing they need.
  • Developer can then send you messages that you have been approved for testing and requesting you download their game.
  • You Sign an online NDA and then start the game.
  • Once they quit the game it takes them to a survey page and they give feedback.

Suddenly with one database programmer and a community manager you could have access to and do coordinated testing with millions of users.

  • With the xbox Arcade type download system tests could get out to testers very quickly.
  • With access to that many testers and stats on them we could do things like control groups and very fine grain testing we have not been able to do before.
  • Given its all online your results are instant and developers can put in code to record all kinds of details you do not get from surveys or just video recorded sessions.  You could even save off a full replay of the session to get reviewed and analyzed later.

Potential issues:

  • Restrictions to ensure it is not abused for marketing purposes.
  • What to do during console transitions and your making a launch title as there are no consoles out yet.  Maybe that just becomes another hardship of being a launch title.
  • Developers will not longer be able to control screenshots and movies during the testing phase.  They can make them sign NDA's but given that is it not done at their building things will end up online.

The potential gains in game quality and developer cost savings are huge as currently we often do not get feedback until it is way to late or too expensive to do anything about it.  I do not know if it would be as easy to implement as I think, but it can not hurt if everyone mentions it to their account rep.  Hopefully someone at Microsoft or Sony will look into it.

 
 
Comments

Adam Bishop
profile image
I don't know if they do it anymore, but during the PS2 days Sony did have semi-public beta tests. I think they were generally limited to a few thousand players, and they were all mostly-finished games being tested for multiplayer balance/stability, but it was definitely something they were doing for a while. I've still got some old beta discs lying around in boxes somewhere of games I tested.

One problem that Sony ran into, which would definitely occur if this were done on a larger scale, was that a lot of console gamers didn't really understand what a beta test was, so they complained pretty frequently about things like unstable code, unbalanced weapons, etc. They didn't realise that that was the whole point of the test to begin with. I think a game could potentially do a lot of damage to its buzz if a bunch of people who didn't understand how beta tests work got access to an early, unstable build and started telling everyone about it.

Travis Johnston
profile image
I remember those beta tests, the costs and the lead times on them were so big that they were hard for the average developer to take advantage of. Since you could not download patches it was a once shot and that limited the kind of testing you could do with it.

Yes there would have to be NDA's and all kinds of things to make sure people understand it is just testing. I agree there may be a little confusion at first but people would get use to it and start providing good feedback as they do want good games to be made. The problem with that Sony system is you usually only usually got one disk for each game so you never saw it improving. With this down loadable system you would often get to see things getting fixed and re-evaluate it to see if it is an improvement. When users see companies listening to them that is the biggest complement and they will in turn start providing massive buzz.

Tomer Chasid
profile image
there's an idea: create software that ports games-in-development to beta-testers via psn, xbla, and net. is this doable?

JB Vorderkunz
profile image
Travis,
I'm a total noob, so please pardon the noobishness of these questions:
Are you suggesting pre-Alpha iteration testing and beyond; Alpha + Beta; or just Beta in the XBL scenario?

Lately, many ppl have been discussing community funding: would there be any possibility of an exemption from the "dev box only" condition for those who have contributed to the game's fund? That is, could they be shipped (or download) the iterations needing testing?

thanks for bearing with me...

Travis Johnston
profile image
Yes I am suggesting this for pre-Alpha iteration testing and beyond.

There is no way to get around the dev box only condition for testing. That hardware manufacturers put in hardware security so that people can not get final disks directly from the developer and mass produce them bypassing their sales. You best bet would be to go to the hardware manufactures and try to get a reduced rate on the dev boxs given that you are trying to do community funding. Your console account rep would be the person to start with.

John Mawhorter
profile image
Even in the PC world where users are more used to bugs and stuff people often take the beta to be the game itself and bad word of mouth can really sink things. This, of course, just means that developers (I'm mostly talking MMO developers, but they are the ones who mostly release betas) only release betas that look really nice, meaning you don't get very significant feedback (because you release the beta later than you would). This is why in-house testers are still the solution in my mind.

Andrew Lockhart
profile image
The focus here seems to be on public focus tests, which are a slightly different creature than gameplay tests. At studios I've worked in my experience has been that plenty of gameplay testing does happen, but that it's restricted to employees in the studio for some of the reasons listed above. In the studio I currently work at we have in-house playtests three times a week at minimum. We're constantly playing and reassessing the state of our game to ensure that a) we're on track in achieving our goals, and b) that we're having fun playing the game. If the answer to either of those is "no" then we make action plans to address any concerns that arise. We have enough of a pool of staff available to ask a handful of people to do a fresh eyes assessment of the game periodically as well. This process starts pre-alpha - the moment we have something that runs, we start testing gameplay.

I'm not suggesting that public focus tests are a bad idea. They certainly have value, but I think it depends on the product and what the goals for the testing are. What kind of information does the team expect to get out of a public test? How will the test environment be controlled?

In addition, if there are studios out there that aren't spending time to play their own game, that's definitely a mistake on their part, and it will show in the final product.


none
 
Comment:
 




 
UBM Techweb
Game Network
Game Developers Conference | GDC Europe | GDC Online | GDC China | Gamasutra | Game Developer Magazine | Game Advertising Online
Game Career Guide | Independent Games Festival | Indie Royale | IndieGames

Other UBM TechWeb Networks
Business Technology | Business Technology Events | Telecommunications & Communications Providers

Privacy Policy | Terms of Service | Contact Us | Copyright © UBM TechWeb, All Rights Reserved.