227 réponses à ce sujet

[Han Shot First]

Rawgrim wrote...

I finish the games several times, though. But it seems to me that most people on this forum has finished alot of Bioware games at least once?

That is because people register on the forums are going to be fans of at least one of Bioware's games, if not more. People who purchased the games but weren't invested in it enough to finish it are unlikely to register or post here. They likely just trade the game in and move on to the next title.

[hoorayforicecream]

Maria Caliban wrote...

How is all that information processed? Does a group just get together after the game goes live and combs it for meaning? Do people search for specific aspects? Do you read reviews and then look at the data to see if any complaints/praise seems to reflect how players behave?

Because it seems like a massive amount of data.

There are servers that take the data and process it as it comes in. Then you have programmers and designers who write scripts to interpret this data in ways that make sense or are useful to them. And yeah, it is a massive amount of data. If you consider that millions of players are generating thousands of data points, you're looking at terabytes of data being collected over the course of days or weeks.

So, for example, you can get a reasonable estimate of the total number of players by counting up the number of unique Bioware accounts that are created and attached to a specific CD key. You can then cross-reference this with the other data associated with the account... have they defeated the Archdemon, whether they visited Orzammar, which followers they have recruited, etc.

The goal is to help provide information to each group on how their work has fared in the real world, and to try to answer the "Why".

Here's an example. Bungie uses their telemetry data for things like multiplayer map balance.

http://www.bungie.ne...97ZEbEX9K5glVE=

The brighter areas on this map are where larger amounts of damage have been dealt to the players playing on the map. As you can see, most of the fighting happens in the outer wings, as opposed to the inner sections or the north section. There are more kills that occur on the east side of the map than the west, suggesting that the map isn't entirely mirrored. The information for this comes from where a player was killed, and knowledge of a specific map. Given enough data points, you can interpret it as a heat map... the bright parts are where the fighting are fierce, and the dark spots aren't. Level designers can then observe games and try to figure out if the decisions they made were the right ones... were they attempting to make the map slightly unbalanced? What is it about the east side that makes it more deadly than the west? What can we use about this information to improve our level designs in the future?

What all of these data points bring is qualification. X accounts have finished the game. Y Shepards romanced Jacob. Z players have replayed the game 23 times to completion. N% of the players have tried more than one origin, but M% of those players never get past Ostagar. If the designers or programmers think the data is important, they can construct queries to find the statistics they want, and use it to help in future development.

[Plaintiff]

BrotherWarth wrote...

Most of DA2's first-week sales came from pre-orders according to Bioware. That means Origins had nothing to do with DA2's lower sales and only benefited its sales. Origins sales climbed as the weeks went on because of the overwhelmingly positive reviews from critics and gamers alike. DA2's sales took a steep dive because of overwhelmingly negative reactions from gamers. Bioware even said as much at the time. DA2 sold poorly because of DA2. It would sold hundreds of thousands fewer units were it not for Origins' stellar status.

Or, an alternative interpretation would be that everyone who cared about Origins and intended to buy DA2 had already preordered, and sales dropped (as is usual) because nobody else was interested.

As to whether or not Origins' success could have been replicated, there's no reason to think it couldn't have been.

Except that virtually no game anywhere has seen a sales pattern like the one Origins experienced. Origin's sales pattern is unique.

ME2 outsold ME1 so why would it have been impossible for DA2 to do as well as Origins?

Unless ME1 and ME2 both had the same unique sales pattern as DA:O, the question is irrelevent. Sequels typically sell better than the original in the first weeks, just like DA2 did. And there's at least one good reason I can think of for why ME2 sold better than ME1, and that's the fact that ME2 was released on all three consoles, whereas ME1 was unavailable to PS3 users until recently.

Origins sold steadily over an increased period of time, which is almost unheard of. And it also had an 'Ultimate Edition', which DA2 did not.

Bioware made a lot of mistakes with DA2 and delivered a sequel of lower quality than its predecessor, plain and simple.

It's not 'simple' at all. Selling well is not an indicator of quality, unless you believe Justin Beiber and the Twilight Saga are the height of modern art.

DA2 lacks mass appeal, apparently, but that speaks nothing of its quality, which is entirely subjective.

Modifié par Plaintiff, 29 novembre 2012 - 08:24 .

[RinpocheSchnozberry]

hoorayforicecream wrote...

[information explosion]

That is pretty damn interesting.  I knew that data was passed back to BioWare/Developers, but I thought it was mostly info on what choices were made in a game, I didn't realize it could be this granular.  This explains the line I saw in a preview along the lines of "BioWare studies Multiplayer to see how people play and have tweaked their design to accomidate that."

Good post!

[Icinix]

I hate seeing analytics used in game development - you can't break down an individual or a groups experience into a set of numbers and graphs.

[Fast Jimmy]

Icinix wrote...

I hate seeing analytics used in game development - you can't break down an individual or a groups experience into a set of numbers and graphs.

It is not inherently bad. But it can be insanely difficult to interpret the right data the right way to make the product better. 

I think the example of MP map level design Hoory just gave is an excellent example. If devs can see how their maps result in most encounters, they can see if that was what they were going for or if there were unintended consequences.

I think the more specific the data, the more relevant. Damage done in aggregate areas of a single map is pretty specific. Number of players who played as a male or as a Human Noble is less so. Average amount of money spent by gamers in microtransactions across the entire industry is pretty much worthless to make any decision off of.

The more "bird's eye view" you get with the data, the less actionable it becomes.

[Icinix]

Oh for sure there are definitely applications for it - but over the last few years I've watched pretty much every industry force and force and force more and more analytical studies / collection into every facet of day to day business and so much of it becomes useless or even detrimental - Entertainment seems to be going down the detrimental path.

[Fast Jimmy]

Icinix wrote...

Oh for sure there are definitely applications for it - but over the last few years I've watched pretty much every industry force and force and force more and more analytical studies / collection into every facet of day to day business and so much of it becomes useless or even detrimental - Entertainment seems to be going down the detrimental path.

It will just make it that much easier for the Robot Overlords to begin running our day-to-day lives with maximum efficiency.


But... a little more on topic... there are many games I have loved over the years, but didn't take the time to hunt down (or sometimes, even take not of) the developer to try and give them feedback, either in the form of "stay the course" or "I didn't like this."

Plus, as my days in analysis of self-report Psychology tests showed me... people are terrible at telling you the truth. Not that people lie, per se, but that they just don't have a really solid grasp on the more mundane details of their existence. People don't count how many times they do something, or how long certain things take, or when they begin feeling certain emotions, both good or bad, during an experience. In that case, you can't really trust people to give you the right data - its better to watch them and see for yourself.

Using a developer's own anecdotal experience, where they say "this worked in this game, this didn't" is something that should never be lost. But, at the same time, many developers are just like fans, where they will say "wouldn't this be cool if feature X were here" or "feature Y would never work" without really understanding where their ideas or concepts are coming from. Sometimes it takes data mining to point out something that should have been obvious from the start.

[zyntifox]

Icinix wrote...

I hate seeing analytics used in game development - you can't break down an individual or a groups experience into a set of numbers and graphs.

As a statistician i recent that comment! Just don't go and tell our clients at the consulting firm i work at that; since we quantify human behaviour daily.

[Maria Caliban]

Icinix wrote...

I hate seeing analytics used in game development - you can't break down an individual or a groups experience into a set of numbers and graphs.

Maybe you can.

(^ Might be NSFW)

Modifié par Maria Caliban, 29 novembre 2012 - 08:54 .

[David Gaider]

Icinix wrote...
I hate seeing analytics used in game development - you can't break down an individual or a groups experience into a set of numbers and graphs.

That's true. You want to avoid drawing erroneous conclusions-- we can know what people did, but not necessarily why they did it. The good thing that this information does is help us not rely solely on anecdotal evidence. We know exactly how many people do what in the game, and that at least informs our decisions if not dictate them.

We know how many people don't complete a game, but just because it's not a lot doesn't mean we shouldn't create an ending or put much effort into it. It does, however, mean we should look at the places where people tend to stop playing and try and figure out why that might be. We know that the vast majority take "good" options, but that doesn't mean we shouldn't include "evil" ones... there might be value placed upon the fact a choice exists even if it's not taken. An individual might do something in a game for any reason, sure, but when you're talking about a sampling pool of this size you can look at events which are statistically significant and act accordingly.

I can see the fear that we might look at this information and discard options which are unpopular-- I think everyone who comes to a forum like this knows, in their heart of hearts whether they admit it or not, that they are far from the norm. The act of coming to this forum alone makes one an anomoly, never mind playing a game repeatedly or arguing on points of nuance for years on end. That's the definition of "hardcore" and that makes you a small, small minority of the audience-- that has always been the case and always will. There are people in the larger audience who might feel similarly about things, neither you nor we know for certain. This kind of information does, however, let us know how they play if not how they feel about it. That, at least, let's us ignore people who try to speak on their behalf regarding things that are factually incorrect (something, quite frankly, people on forums like to do a great deal)... but doesn't mean that we'll automatically also ignore opinions on things people have every right to have opinions on. That's also feedback, and it's useful.

If one's concern, however, is that a developer should only be listening to the hardcore audience because they're the ones willing to speak their minds, and ignore factual data because it might indicate most people don't play the game the way they do... well, that has never happened. Ever.  This simply makes us better-informed, nothing more.

[rapscallioness]

Wow. That is some interesting info about data collection.

I wonder if the ME team used this kinda thing as reason for trying to implement so much auto dialogue? They said something about how it was too pause-y before. It didn't flow. And that they think w/auto-d it flows better.

I wonder if they were trying to interpret stats of players just standing there during convo.s. Idle. and thought that maybe the player was bored, or something? Maybe they thought to keep the player involved by flowing the scene up?

Not taking into account that some players are paused cuz they are thinking about what they want to say. Or, they went to make a sandwich/use the toilet/crack open a beer...or other party favors. Then settle back in for a nice convo break. Or, even just admire their PC creation?

Cuz for me it would be the exact opposite. The prevalence of auto-d bores the hell outta me. I'm just not as involved in the convo cuz..well, because I'm not actually involved in the convo.

I also wonder if this had anything to do w/the decision to take out the neutral dia. choices? Thinking no one was picking the neutral choices so...scrap them? When, of course, with that scenario it's more cuz of the "penalty" for picking neutral. Or even vacillating between para/ren.

Unless the game was ng+, lil pixel lives were being risked everytime the player picked neutral, or vacillated.

I have no idea if this kinda data played a role in those choices from the ME team, but I do wonder.

[Mr Fixit]

Maria Caliban wrote...

Maverick827 wrote...

This thread exists in some weird void where haters/RPG Elitists attack the OP because they think he's saying that games should be built for the "silent majority" and not for them, and fanboys/casuals attack the OP because they think he's saying that games are being made for the silent majority and that they don't deserve it.

It seems like both sides are engaging in some sort of "who can take the most misinterpreted offense" contest.

Your post seems to exist in a weird void where Viktoria is a man's name.

Ever seen Victor Victoria?

Modifié par Mr Fixit, 29 novembre 2012 - 09:17 .

[Han Shot First]

I wonder how large of a percentage of those people who don't complete the games, don't do so simply because they aren't really into RPGs.

Sometimes people buy games that may be outside of what they usually play, simply because the game has great word of mouth. They hear it is awesome enough times to give it a try, only to find out it isn't for them. Or at least that has been my admittedly limited experience with people who purchase DAO or Mass Effect. Out of my friends I have only one other that enjoys RPGs, the rest are primarily into shooters. I know a couple of the latter who bought DA:O or one of the ME games, and never completed them. One of them was playing ME3's multiplayer for awhile, but never got beyond Sur'Kesh (an early mission) in the single player game.

Modifié par Han Shot First, 29 novembre 2012 - 09:20 .

[Fredvdp]

David Gaider wrote...

We know that the vast majority take "good" options, but that doesn't mean we shouldn't include "evil" ones... there might be value placed upon the fact a choice exists even if it's not taken.

This describes my feelings towards Mass Effect 2. I rarely choose a neutral option, but I'm glad they exist. Same with investigate options. I sometimes wondered if the Mass Effect devs noticed that players rarely pick neutral options and therefore decided to get rid of them in ME3. I didn't like that at all.

The reason I never picked neutral options is because you were punished for not maxing either paragon or renegade in ME2. In ME3 you're allowed to be neutral, so it makes no sense that they got rid of the neutral choices.

Modifié par Fredvdp, 29 novembre 2012 - 10:00 .

[Fredvdp]

Han Shot First wrote...

I wonder how large of a percentage of those people who don't complete the games, don't do so simply because they aren't really into RPGs.

I have a brother who always buys games without doing any research. He played the Gears of War 2 tutorial and then sold it because he realized he doesn't like third person shooters.

Modifié par Fredvdp, 29 novembre 2012 - 09:56 .

[AlanC9]

rapscallioness wrote...
I wonder if the ME team used this kinda thing as reason for trying to implement so much auto dialogue? They said something about how it was too pause-y before. It didn't flow. And that they think w/auto-d it flows better.

I wonder if they were trying to interpret stats of players just standing there during convo.s. Idle. and thought that maybe the player was bored, or something? Maybe they thought to keep the player involved by flowing the scene up?

Well, to hear Patrick Weekes talk -- he's about the only ME guy who's been really forthcoming lately --it was more about cutting out the choices that didn't change anything in favor of better flow in the convos. Autodialogue was intended to replace fake interaction that always took you to the same place no matter what order you hit the investigate choices in, if you hit them at all. I'm replaying ME2 right now, and I can see a lot of places where interacting with the wheel is just busywork.

Which doesn't mean that ME3 didn't take the approach too far. I presume DA3 won't make that particular mistake again.

Modifié par AlanC9, 29 novembre 2012 - 10:16 .

[Bernhardtbr]

Icinix wrote...

Oh for sure there are definitely applications for it - but over the last few years I've watched pretty much every industry force and force and force more and more analytical studies / collection into every facet of day to day business and so much of it becomes useless or even detrimental - Entertainment seems to be going down the detrimental path.

The problem is very simple, remove statistics and you are left with personal opinion to do your analysis, which is often MUCH more biased and faulty.

Good statistics helps a lot. For example, if medicine has so many good treatments, instead of people still saying that tea or ****** can cure flu, it´s due to statistics and decent clinical trials.

However, as very well put before, the problem with statistics is that they often show what happens, and not WHY it happens.

Modifié par Bernhardtbr, 29 novembre 2012 - 10:29 .

[Nerevar-as]

AlanC9 wrote...

rapscallioness wrote...
I wonder if the ME team used this kinda thing as reason for trying to implement so much auto dialogue? They said something about how it was too pause-y before. It didn't flow. And that they think w/auto-d it flows better.

I wonder if they were trying to interpret stats of players just standing there during convo.s. Idle. and thought that maybe the player was bored, or something? Maybe they thought to keep the player involved by flowing the scene up?

Well, to hear Patrick Weekes talk -- he's about the only ME guy who's been really forthcoming lately --it was more about cutting out the choices that didn't change anything in favor of better flow in the convos. Autodialogue was intended to replace fake interaction that always took you to the same place no matter what order you hit the investigate choices in, if you hit them at all. I'm replaying ME2 right now, and I can see a lot of places where interacting with the wheel is just busywork.

Which doesn't mean that ME3 didn't take the approach too far. I presume DA3 won't make that particular mistake again.

I don´t think the point was that Shepard talked too much on his/her own, raher than it ended showing emotion without the player input. I remember being really annoyed by the Tessia aftermath, by Shepard being depressed when I wanted him angry and frustrated.

I´m also a bit worried by BW tending towards cinematography. I´m playing a RPG, not watching an interactive movie.

[Kidd]

hoorayforicecream wrote...

*telemetry info drop*

*jawdrop* Thanks for posting that, I had no idea things could be just that advanced. That looks incredibly cool and useful! =) Glad telemetry can be used like that.

David Gaider wrote...

I think everyone who comes to a forum like this knows, in their heart of hearts whether they admit it or not, that they are far from the norm. The act of coming to this forum alone makes one an anomoly, never mind playing a game repeatedly or arguing on points of nuance for years on end.

^People, don't read this part while listening to depressing music. The words start meaning something else entirely! *switches song, rereads post* There, much better~ =)

BrotherWarth wrote...

That's another nonsensical argument. No one was buying the Collector's Edition and standard edition of Origins. And DA2 had the Signature Edition.

Speak for yourself [smilie]http://social.bioware.com/images/forum/emoticons/wizard.png[/smilie] I picked up a CE, SE and an UE alike - all for different platforms. I honestly don't think it was all too rare for people to pick up an UE afterwards if they were big fans, yet not hard-hardcore fans who bought all the DLC as it came out.

[zyntifox]

I've got two standard edition of DA:O, one for my PS3 and one for my MAC. Never did play the one for my MAC because... well 13-inch screen vs 50-inch screen. Later bought the UE to the PS3, obviously after i bought most of the DLC to my standard editions.

[Foolsfolly]

Fredvdp wrote...

David Gaider wrote...

We know that the vast majority take "good" options, but that doesn't mean we shouldn't include "evil" ones... there might be value placed upon the fact a choice exists even if it's not taken.

This describes my feelings towards Mass Effect 2. I rarely choose a neutral option, but I'm glad they exist. Same with investigate options. I sometimes wondered if the Mass Effect devs noticed that players rarely pick neutral options and therefore decided to get rid of them in ME3. I didn't like that at all.

The reason I never picked neutral options is because you were punished for not maxing either paragon or renegade in ME2. In ME3 you're allowed to be neutral, so it makes no sense that they got rid of the neutral choices.

I never got the whole 'punished' thing with ME2. The two biggest Influence checks in the game were the Tali/Legion and the Miranda/Jack moments on the Normandy. Both of which had non-high Influnece paths to keep both loyal.

Lair of the Shadow Broker added another one but there's still two different non-morality influence options that end in different but essentially the same out-comes. It's really no different than being 'punished' for not having any skill up to snuff during your garden variety skill-check.

[Realmzmaster]

I bought the Ultimate edition of DAO because I did not have all the dlc. I already had the CE for DAO. If I had purchased all the dlc separately it would cost more than the UE. I also have Awakening as a separate expansion and two copies of DA2 Signature edition. So yes gamers buy the UE even if they have the CE and some of the dlc. It simply a matter of economics.

I just recently bought the Game of the Year edition of Fallout 3 for $9,95. I like the game but not enough to buy it when it first appeared and then buy the dlc. I bought Fallout New Vegas for $9.95. I will purchase the Deluxe edition when the price drops further to get the dlc.

[Foolsfolly]

Realmzmaster wrote...

I bought the Ultimate edition of DAO because I did not have all the dlc. I already had the CE for DAO. If I had purchased all the dlc separately it would cost more than the UE. I also have Awakening as a separate expansion and two copies of DA2 Signature edition. So yes gamers buy the UE even if they have the CE and some of the dlc. It simply a matter of economics.

I just recently bought the Game of the Year edition of Fallout 3 for $9,95. I like the game but not enough to buy it when it first appeared and then buy the dlc. I bought Fallout New Vegas for $9.95. I will purchase the Deluxe edition when the price drops further to get the dlc.

In general, I don't grab them.

DA:O, however, for what ever reasons is the one time I can point and say otherwise. I own two DA:Os, DA:A, bought two of the DLCs, and then bought the Ultimate Edition entirely for the rest of the DLC.

...

...Otherwise I just pick up the DLC as it comes out or in Fallout 3's case I really did wait for the Ultimate edition to come out to pick the game up.

I may pick up the New Vegas Ultimate one day. Maybe. I have all the NV DLC but I have to admit I like having hard copies of DLC. I don't know why. I also like paper books and comics more than digital comics or books.