Jump to content

The Best View in SWTOR contest has returned! ×

WHY does it take so long to FIX something ? 6.0 is my last hope...


DavidAtkinson

Recommended Posts

In my opinion as a loyal fan of the game since march 2015, I can safely say that 5.10 is the worst update released for this game since I am playing and it's probably the worst update released since launch....

 

What's the point of the PTS if developers don't listen to feedback ? Like why should any player even bother to say something when bioware was told that set bonuses are bugged, mods and enhancements are bound to slots and it still gets released like that to ruin our min maxing of stats.

 

5.10 is full of game breaking bugs that seems like they take forever to get fixed and I'm sure any other studio would have fixed them pretty quick with a patch. But not Bioware.. and I am having trouble understanding why is this game so damn slow. It takes forever to fix something.. there are bugs that are in the game for years and still haven't been even talked about. Is the team working on this game really that small that releasing patches is that hard ? I

 

For me as a player SWTOR is becoming slowly a waste of money and time game as it isn't moving forward at all. We've had 2 years of total stagnation and here we are now with a HUGGE silence regarding this game's future and with a 6.0 that may or may not come....

 

For me 6.0 is the last hope for this game since it might be an expansion sized update which means some things can be improved, REAL CONTENT can be added and not some boring 10 minute story like Ossus.

 

The direction this game has taken since KOTET is depressing since I fail to remember anything of note released since then. We got very small chunks of story PVP maps that everyone hates(vandin).

 

I still remember Bioware saying that the time of large expansions is long gone and they believe that releasing small content over time is a better idea.

 

Well, this has me a bit worried about 6.0 but I just cannot see it being even half the size of vanilla 5.0.

 

So when will it be released ? I think if the don't get any news until may/early chances are it isn't happening in 2019....

 

As for myself, I am very tempted to look elsewhere since it seems like other MMO"s like ESO are treated much better by their developers.

Link to comment
Share on other sites

I've been playing since day 1 (not actively all the time; throughout 2016-2018 I mostly did not play at all) and there are bugs from classic SWTOR still in the game. But the worst bugs came in 2.0, in my opinion. The engine can't seem to handle abilities such as Exfiltrate. Players have a habit of disappearing from the screen for a couple of seconds after they've used that ability, especially when they use it to leap of ledges. They've not taken any steps to fix that, and it's such a big glaring issue! It's not just annoying but it looks totally broken. There's rubberbanding in PvP since classic SWTOR which is very apparent after you've been force pushed. I made this video in 2013: https://youtu.be/A6oRTlRdC-c ... and it is still like that whenever I get force pushed. Does that look like good game feel?
Link to comment
Share on other sites

Considering that it feels like they're down to 1.5 devs working on SWTOR, don't hold your breath.

They need time to fix what they broke, first, before layering more broken content on top of existing broken content.

 

Rushing 6.0 won't beget playable content.

 

Ever since 5.0 was launched a lot of damage was done to this game.. starting with the stupid command system and most of all with the lack of content/ forcing players to do content they hate... and the gearing system that was improved and now with 5.10 it's another absurd gearing. No one minds gearing.. but now they've gone way to far.

 

All these things considered the game is doing decently if I look at the population numbers at least on DM server.. pretty good. Seems like the star wars name is pretty strong to keep this game going despite all... but It could be so much better.

Edited by DavidAtkinson
Link to comment
Share on other sites

The main reasons for not only the introduction of bugs but the time to fix them for this game is this:

 

* Poor in-house testing. Specifically, no established regression suite. (There are various ways to determine this, even from the outside looking in. If such a regression suite does exist then it either isn't being run, its results aren't being paid attention to, or it is written in such a way that data conditions are not varied enough.)

 

* No use of an orchestrator type tool, the use of which is very standard in MMO testing. (This is known because of the type of bugs we see but also because the Hero Engine is not, and has never been, amenable to orchestration to the extent that it could be.)

 

* Very tightly coupled code. Too much code is interrelated to other code, causing side-effect bugs. (This is demonstrable any time you have certain changes but then bad effects seem to occur in functionality that seems "far apart" from what was changed. A good recent example is the disappearing faces from some body types.)

 

* Lack of listening to feedback from PTS or, at best, very selective listening of feedback. This leads to problems introduced that eventually become part of the backlog.

 

* A severe backlog of issues that was allowed to build up over time. This led to bugs not getting fixed because there were too many. This in turn led to a lot of functionality being built on top of bugs, such that any fixes now risked breaking other functionality. Without the regression suite or the orchestration I mentioned, this becomes even more problematic to test and thus harder to verify in terms of fixing.

 

I wrote up some examples of this testing problem with SWTOR in particular here and here and here and, finally, here.

Link to comment
Share on other sites

The main reasons for not only the introduction of bugs but the time to fix them for this game is this:

 

* Poor in-house testing. Specifically, no established regression suite. (There are various ways to determine this, even from the outside looking in. If such a regression suite does exist then it either isn't being run, its results aren't being paid attention to, or it is written in such a way that data conditions are not varied enough.)

 

* No use of an orchestrator type tool, the use of which is very standard in MMO testing. (This is known because of the type of bugs we see but also because the Hero Engine is not, and has never been, amenable to orchestration to the extent that it could be.)

 

* Very tightly coupled code. Too much code is interrelated to other code, causing side-effect bugs. (This is demonstrable any time you have certain changes but then bad effects seem to occur in functionality that seems "far apart" from what was changed. A good recent example is the disappearing faces from some body types.)

 

* Lack of listening to feedback from PTS or, at best, very selective listening of feedback. This leads to problems introduced that eventually become part of the backlog.

 

* A severe backlog of issues that was allowed to build up over time. This led to bugs not getting fixed because there were too many. This in turn led to a lot of functionality being built on top of bugs, such that any fixes now risked breaking other functionality. Without the regression suite or the orchestration I mentioned, this becomes even more problematic to test and thus harder to verify in terms of fixing.

 

I wrote up some examples of this testing problem with SWTOR in particular here and here and here and, finally, here.

 

This post above gives some really good explanations as to why the game is broken every time they add a big patch, I never knew the technical parts to it so much but it makes total sense to me as explained by Kryptonomic.

 

All of this is why it's astonishing that they keep dismantling parts of the game that has been fine tuned and developed only to gut it and replace those parts with entirely different working systems.

 

They did it with gearing, multiple times now. They did it with conquest. Why gut and destroy parts of the game which are working fine (usually after being tweaked and fixed for months) only to replace them with untested faulty parts?

 

Why would they not just continue to build on top of successful working pieces on the game that have already been thoroughly tested over time and found to be reliable as well as enjoyable to the players?

 

The most frustrating thing is, the majority of the changes were never needed, never asked for, and almost always unpopular. It's as if the SWTOR management have no idea what the actual players enjoy about the game. They remove most of the things that make the game fun!

 

It's just mind boggling. Supposedly they are working with a limited crew of devs, limited funds, yet this is how they allocate the funds and dev work?

 

Top the sundae off with awful communication, misdirection, and hidden agendas and you get nothing but a really rotten cherry on top.

 

As time has gone on, as noted by Kryptonomic the game has gotten worse bug-wise with each large patch and his explanation of how the coding is done paints the picture as to why one change can end up bugging unrelated areas of the game. I found that part interesting because I never knew that.

 

Thanks to Krpytonomic for going into detail on his take on what has happened and is happening to the game. Makes more sense to me now.

Edited by Lhancelot
Link to comment
Share on other sites

All of this is why it's astonishing that they keep dismantling parts of the game that has been fine tuned and developed only to gut it and replace those parts with entirely different working systems.

 

Some of this has to do with that tight coupling. For example, without getting into all the boring details, there is a certain level of mapping that exists between the databases that hold resources and assets and the game engine itself. So let's take one example from the past.

 

At one point, companions were limited in the types of role they could take. Mako was only a healer, for example. Khem Val was only a tank. This was a mapping in the database. Another mapping was the "combat dialogue" that each companion would make. When the companions were updated such that they could take any role, the tight coupling meant that sometimes the "combat dialogue" would get associated with the wrong companion or would become null in the database.

 

In the latter case, this leads to companions that are silent. In the former case, you might have Quinn or Kira sounding like someone else during combat.

 

Why would they not just continue to build on top of successful working pieces on the game that have already been thoroughly tested over time and found to be reliable as well as enjoyable to the players?

 

Therein lies the problem. The parts that are successfully working are coupled to other parts. Sometimes when one of those other parts goes bad, the previously working parts goes bad as well.

 

The most frustrating thing is, the majority of the changes were never needed, never asked for, and almost always unpopular. It's as if the SWTOR management have no idea what the actual players enjoy about the game. They remove most of the things that make the game fun!

 

I realize this opinion will differ among people but I do believe it's at least fairly objective to say that the original development team was a bit more capable around the ideas of RPGs and MMOs in particular. (You can look at LinkedIn both now and in the past for who was associated with the game and based on that and the experience listed, make some determinations.) The team that ended up working on the game after the major development turnover -- which was very early in SWTOR's history -- did not have that same level of skill.

 

I'm not saying they are "bad programmers" or don't care about their jobs or anything like that. But I do know they were given a modified engine to work with. The original engine itself wasn't the best documented and the modifications made to it by the alpha team were opaque enough to mean that changes had to be handled with care, at least when you got into the core mechanics.

 

Couple this with the fact that it rapidly became clear that the very early original approach of KOTOR 3 is what many fans wanted; they didn't want an MMO. But that ship had sailed. So now a lot of story elements, such as relationships and complex flags in the database indicating who died, who lived, etc, were now sitting side by side with mappings not related to story. Yet -- again -- tight coupling means the one can impact the other. Hence we get companions that can't be summoned; or people who were killed that are now alive again somehow.

 

Top the sundae off with awful communication, misdirection, and hidden agendas and you get nothing but a really rotten cherry on top.

 

I sympathize with your viewpoint here. Having worked with many community management teams in the context of testing games, I can say that the SWTOR team definitely is a bit inexperienced in this regard.

 

They are very passively defensive, quick to shut down certain discussions they don't like. That often happens with inexperienced teams when things are happening that a lot of players don't want to hear. Many long-standing questions are simply ignored. Even if there is no good answer, any community manager will tell you it is better to at least state that rather than say nothing.

 

Most pernicious, I feel, is a simple lack of anticipating what players will want to hear. A good example was the recent initial announcement of the "back-end only" server maintenance. Any community team should have known that with the amount of bugs introduced in 5.10, mention should have been made of that. Something like: "We know there are a lot of bugs that need to be dealt with and I'm sure you are disappointed to hear that our first maintenance isn't about fixing those. But here's what we are aware of and while we want them fixed just as fast as you probably do, we also want to make sure we get this as right as possible."

 

You get the idea. I'm not trying to write their material for them but just that little bit of anticipation about how players will react and thus crafting the message accordingly can go a long way.

 

Right now -- at least looking from the outside in -- the SWTOR team is in a moderately uncomfortable spiral of pressure regarding having to add more (to satisfy content demands) but with much more limited resources than they have had in the past. This coupled with the legacy of their bug backlog and some of the architecture choices of the alpha team puts them in an untenable position in many cases. I'm not saying that to excuse anything necessarily but it is also a context to be aware of.

Edited by Kryptonomic
Link to comment
Share on other sites

Some of this has to do with that tight coupling. For example, without getting into all the boring details, there is a certain level of mapping that exists between the databases that hold resources and assets and the game engine itself. So let's take one example from the past.

 

At one point, companions were limited in the types of role they could take. Mako was only a healer, for example. Khem Val was only a tank. This was a mapping in the database. Another mapping was the "combat dialogue" that each companion would make. When the companions were updated such that they could take any role, the tight coupling meant that sometimes the "combat dialogue" would get associated with the wrong companion or would become null in the database.

 

In the latter case, this leads to companions that are silent. In the former case, you might have Quinn or Kira sounding like someone else during combat.

 

 

 

Therein lies the problem. The parts that are successfully working are coupled to other parts. Sometimes when one of those other parts goes bad, the previously working parts goes bad as well.

 

 

 

I realize this opinion will differ among people but I do believe it's at least fairly objective to say that the original development team was a bit more capable around the ideas of RPGs and MMOs in particular. (You can look at LinkedIn both now and in the past for who was associated with the game and based on that and the experience listed, make some determinations.) The team that ended up working on the game after the major development turnover -- which was very early in SWTOR's history -- did not have that same level of skill.

 

I'm not saying they are "bad programmers" or don't care about their jobs or anything like that. But I do know they were given a modified engine to work with. The original engine itself wasn't the best documented and the modifications made to it by the alpha team were opaque enough to mean that changes had to be handled with care, at least when you got into the core mechanics.

 

Couple this with the fact that it rapidly became clear that the very early original approach of KOTOR 3 is what many fans wanted; they didn't want an MMO. But that ship had sailed. So now a lot of story elements, such as relationships and complex flags in the database indicating who died, who lived, etc, were now sitting side by side with mappings not related to story. Yet -- again -- tight coupling means the one can impact the other. Hence we get companions that can't be summoned; or people who were killed that are now alive again somehow.

 

 

 

I sympathize with your viewpoint here. Having worked with many community management teams in the context of testing games, I can say that the SWTOR team definitely is a bit inexperienced in this regard.

 

They are very passively defensive, quick to shut down certain discussions they don't like. That often happens with inexperienced teams when things are happening that a lot of players don't want to hear. Many long-standing questions are simply ignored. Even if there is no good answer, any community manager will tell you it is better to at least state that rather than say nothing.

 

Most pernicious, I feel, is a simple lack of anticipating what players will want to hear. A good example was the recent initial announcement of the "back-end only" server maintenance. Any community team should have known that with the amount of bugs introduced in 5.10, mention should have been made of that. Something like: "We know there are a lot of bugs that need to be dealt with and I'm sure you are disappointed to hear that our first maintenance isn't about fixing those. But here's what we are aware of and while we want them fixed just as fast as you probably do, we also want to make sure we get this as right as possible."

 

You get the idea. I'm not trying to write their material for them but just that little bit of anticipation about how players will react and thus crafting the message accordingly can go a long way.

 

Right now -- at least looking from the outside in -- the SWTOR team is in a moderately uncomfortable spiral of pressure regarding having to add more (to satisfy content demands) but with much more limited resources than they have had in the past. This coupled with the legacy of their bug backlog and some of the architecture choices of the alpha team puts them in an untenable position in many cases. I'm not saying that to excuse anything necessarily but it is also a context to be aware of.

 

I appreciate the time you put into this post. Found it enlightening and objective.

 

Yeah, my opinion is not well met by some on the forums and I probably don't hide my disappointment or resentment well when discussing the problems I have with the management of this game.

 

I agree with everything you said regarding how they manage their communication, I never could have put it that way, but it's so spot on, the part about them being "passively defensive" and how you mentioned how they have a "lack of anticipating what players will want to hear."

 

I always refer to this as them not having their finger on the pulse... But your wording is more concise and accurate.

 

Thanks for sharing your perspective. Gives me some peace knowing I aint crazy in my criticisms but also the technical aspects you explained about the coding was telling and gives a reasonable explanation to how and why patches end up causing so many bugs.

Edited by Lhancelot
Link to comment
Share on other sites

That's cute... you really think that IF we get a 6.0 (yes, IF, there's never been any guarantee of such), it won't be exactly as (if not even more) broken and bug-riddled as every other update EA has put out these past couple of years? This is the company that keeps putting stuff on the PTS or in closed betas and then ignores every single bit of feedback they receive and that seems allergic to bug testing any SWTOR code. We'll be lucky if the servers even stay up on the theoretical day 6.0 drops. Edited by AscendingSky
Link to comment
Share on other sites

<<takes sip of martini and dons flame retardant White Knight cloak>>

 

We are all entitled to define "game breaking" individually. We each have our own threshold.

 

Personally, I define "game breaking" as ruining the ability to play. They fixed the Jaesa bug almost immediately, which was game breaking for those who raised her (like I did) on a character almost immediately.

 

I can live with the occasional headless character; I can live with the Jedi Artifact quest only working sometimes, and for no apparent reason.

 

<<takes sip of martini and switches to Grey Knight cloak>>

 

The companion AI bug is irritating, immensely so, but it just requires a bit more micro-managing. Annoying as heck, but hardly game breaking.

 

I completely agree with everything Krypto posted. I can do so and also call out the OP for being overly dramatic, particularly since s/he did not specify the bugs that are game breaking. The two posts are not mutually exclusive.

 

<<takes sip of martini, goes back to playing, and sees the servers more populated than any time in recent memory>>

 

Dasty

Edited by Jdast
Link to comment
Share on other sites

We are all entitled to define "game breaking" individually. We each have our own threshold.

 

Absolutely true. In the software development industry as a whole, not just game development, a bug is technically defined as "anything that threatens value." Obviously an outright crash or the lack of ability to progress a story would be an immediate threat to the value of most players.

 

It's equally true to say that quality is a shifting perception of value over time. It can be dependent upon such things as emotion and mood. (That jumping mechanic to get a datacron can be really fun when you have a lot of time on your heads; it can seem like the worst buggy design in the world when you simply are trying to get off planet as quickly as possible.)

 

Taking these points together, this means that people have exactly what you stated: different thresholds for what they are and are not willing to tolerate in terms of bugs. That's the case regardless of whether the bugs are considered game-breaking or not.

 

Speaking broadly here for a second, MMO development in the industry is widely known to be one of those areas where it's both hardest to please the majority of players and yet also the easiest. MMO players (again, speaking broadly here) are well-known for accepting highly variable quality but that's also because they are more willing to run content over and over again. MMO players often want content-content-content, sometimes regardless of the bugs -- as long as the bugs don't stop the gear grind.

 

Story-driven players, however, are much less accepting of variable quality, particularly if it ruins their immersion. This is even more so if the game in question is based on a lore-heavy franchise, whether from books, films, or just from its own established brand.

 

Now consider a game like SWTOR which is clearly an MMO in many ways but equally clearly can be played entirely as a single player game; at least for the story content. Now consider that the single-player RPG story is situated in the MMO. Or wait. Is it that the MMO is situated in the single-player RPG?

 

SWTOR development has tried to balance both to greater and lesser degrees. And that schizophrenic development cycle can play havoc with some of the other things we've talked about in this thread, but notably the prioritizing of bugs.

 

And therein lay a lot of the quality problems that people have perceived over time with SWTOR, but often from very different perspective and vantage points. And when that happens, of course, you have a player base that can often speak past each other (and note that's without having to say one side is "right" or "wrong") and thus a community team that becomes singularly entrenched in the passivity of non-communication because -- as they find -- any communication can lead to vast swathes of their player-base being unhappy about one point or another.

 

Obviously I'm generalizing quite a bit here and while I don't think it's throwing off my accuracy stat too much (see what I did there?), it's certainly important to at least be aware of it, I guess.

Edited by Kryptonomic
Link to comment
Share on other sites

<<takes sip of martini and dons flame retardant White Knight cloak>>

 

We are all entitled to define "game breaking" individually. We each have our own threshold.

 

Personally, I define "game breaking" as ruining the ability to play. They fixed the Jaesa bug almost immediately, which was game breaking for those who raised her (like I did) on a character almost immediately.

 

I can live with the occasional headless character; I can live with the Jedi Artifact quest only working sometimes, and for no apparent reason.

 

<<takes sip of martini and switches to Grey Knight cloak>>

 

The companion AI bug is irritating, immensely so, but it just requires a bit more micro-managing. Annoying as heck, but hardly game breaking.

 

I completely agree with everything Krypto posted. I can do so and also call out the OP for being overly dramatic, particularly since s/he did not specify the bugs that are game breaking. The two posts are not mutually exclusive.

 

<<takes sip of martini, goes back to playing, and sees the servers more populated than any time in recent memory>>

 

Dasty

 

There have been reports made by people who's game crashes every time they try to load in. That's game breaking. There are people reporting that key quest dialogues won't trigger right so they can't progress. That's game breaking. There are other companions that suffer from the same fate Jaesa did, so by your own definition, that's game breaking. For story players, having the character flags jumbled up is also game breaking. There are a host of other bugs, that while I don't consider them game breaking, are awfully bad, like GSF wins not counting for the weekly. For decorators, having numerous decorations broken is pretty bad. For me, losing my new character's face and having to pay CC to do a work around was pretty bad. I have never in 20 years of playing MMOs had to spend real money to do a work around for a bug before.

Link to comment
Share on other sites

<<takes sip of martini, goes back to playing, and sees the servers more populated than any time in recent memory>>

 

Dasty

 

Well heavily populated servers is easy to achieve if you keep consolidating the player-base on to ever fewer servers.

 

 

All The Best

Link to comment
Share on other sites

<<takes sip of martini and dons flame retardant White Knight cloak>>

 

We are all entitled to define "game breaking" individually. We each have our own threshold.

 

Personally, I define "game breaking" as ruining the ability to play. They fixed the Jaesa bug almost immediately, which was game breaking for those who raised her (like I did) on a character almost immediately.

 

I can live with the occasional headless character; I can live with the Jedi Artifact quest only working sometimes, and for no apparent reason.

 

<<takes sip of martini and switches to Grey Knight cloak>>

 

The companion AI bug is irritating, immensely so, but it just requires a bit more micro-managing. Annoying as heck, but hardly game breaking.

 

I completely agree with everything Krypto posted. I can do so and also call out the OP for being overly dramatic, particularly since s/he did not specify the bugs that are game breaking. The two posts are not mutually exclusive.

 

<<takes sip of martini, goes back to playing, and sees the servers more populated than any time in recent memory>>

 

Dasty

 

If this doesn't put things into perspective as far as how "great" the population of SWTOR is now, I don't know what will.

 

https://www.reddit.com/r/swtor/comments/372c3j/i_just_found_out_there_was_once_this_many_servers/

 

It's fine to point out when people say "game is dead" that they are wrong, but come on Dasty, let's use context here and be honest.

 

The game has drastically regressed in population over time and only losing more as the game suffers awful design changes that come with the added bonus of baked in bugs affecting every facet of the game. Pretending the game's population is thriving and the servers are better than ever is disingenuous at best.

 

Also noting that certain bugs for you are game breaking while those listed by others not-so-much is a bit rich. You clearly realize "game breaking" is subjective depending on what activities a player enjoys most on SWTOR and what bugs are running rampant as you even said yourself... So when someone else claims it's game breaking for them, it is.

Link to comment
Share on other sites

I'm going to consolidate my responses because I don't want to be like Papa Smurf (ooh that was good :rak_03:)! You know who I mean.

 

To Lhance and Spudsie (can I call you that? :rak_03:),

 

My baseline for comparison was after the server mergers took place. I'm not a moron, I started out on Jung Ma in 2011 and have witnessed (albeit after a three year break when I didn't know the game still existed and rejoined in Dec. 2016) the consolidation. In this case, there has been an uptick compared to the last year. Hardly a surprise, since it is the first story content since the FP Crisis on Nathema (May 2018). That's all I'm saying.

 

To Damask, it's hard to reply to what you said because my game loads. Obviously, I would not pay for a game that doesn't load. As for the missing companions, yup, I agree. In the case when Jaesa was level 1 (due to bug), I played an alt. I wrote that was game breaking because I had alternatives to play since I'm an altoholic but recognized others did not. (Ditto with Darth Hexid and Paxton bug).

 

To Krypto, you neither desire much less need my kudos -- but I think every post you made in this thread has been awesome. Well articulated, explained. I say that as someone in a former management who screamed at IT to get things done. Apparently (to show I can be self-deferential), one of the worst things you can say to an IT person, which I did, is: "We landed a S@()#%*Nsfkjfs guy on the #)$U)@($ULK Moon in 1969, why can't you get our severs up?" FYI, that was learning experience. :rak_03:

 

Look, anyone who has played this game for any length of time knows that we are beta testers for a patch release. My only point is that there is a categorical (if not by degree) difference between game breaking and annoyance.

 

Peace,

 

Dasty

Edited by Jdast
Link to comment
Share on other sites

We are all entitled to define "game breaking" individually. We each have our own threshold.

 

That threshold can be crossed by repeated exposure to the same bugs, though. So while a bug may not be immediately game breaking, it may still be the cause why someone leaves the game. Certain bugs I can live with; for instance, the sound effect on Force in Balance has been broken since sometime around 2.0. I sent a ticket, they said they would investigate and nothing has happened. I can live with that bug. The rubberbanding issue that I was talking about in my previous post, however... that is what kills the game for me after repeated exposure. Especially when they add new abilities that the game engine can't handle which just makes the issue all the more present and frustrating. Take Phantom Stride and Holotraverse, as an example. In PvP, how often do these new abilities actually get you to your target and how often do they fling you up into the air or some other nonsense like that?

 

These are shiny, new and cool abilities that the game engine ultimately can't handle. One might wonder why they even put them in the game to begin with, knowing what their engine is incapable of.

Edited by Majspuffen
Link to comment
Share on other sites

These are shiny, new and cool abilities that the game engine ultimately can't handle. One might wonder why they even put them in the game to begin with, knowing what their engine is incapable of.

 

Speaking to this point alone, this is exactly why orchestrators, which I mentioned somewhere upthread, are used in games like this for testing purposes. It's to see not only that something works, but checking whether that something has a quality degradation under varying conditions (i.e., rendering, load, environment priority bands, etc).

 

An interesting tidbit, perhaps, is that sometimes the testing done on development shards will reflect the abilities working just fine. It's when the functionality gets put onto production instances that things start to go a bit haywire. This is why public test servers are so important, assuming the public test server has good enough fidelity with the production environment. (Ideally, they are equivalent to the production server in terms of configuration and load possibilities.) And, of course, assuming that the development team is aware of and listening to feedback from those participating in the public testing.

Link to comment
Share on other sites

The question is... why Bioware released a new tier 5 gear system just before vacation... wasn't easiest to simple release Ossus with out the Tier 5 Gear update ?

 

More than likely this was simply a case of the work they committed to having done in their equivalent of the development sprint. If internal project commitments are made, it's considered problematic if those commitments aren't met.

 

For example, if work has been planned out for various sprints, then not reaching each sprint goal puts upcoming sprints in danger of not being able to achieve their goals because now work from the previous sprint has to be incorporated. And so on and so forth.

 

Now, of course, it's still possible to ask: "Well, then why was the Tier 5 Gear upgrade allocated for the same sprint as the Ossus content?" That we can't have an answer to without more insight into the project discussions at the time and the items that are planned for their upcoming sprint cycles.

 

One other note that may be of interest for those who don't work in the development industry (game or otherwise): lots of bugs are a huge friction on team velocity. (Team velocity being the time it takes for the development team to deliver features of an agreed upon quality within an agreed upon timeframe.)

 

So that's the interesting part of the development industry: sometimes we rush to meet our commitments for sprints -- but introduce tons of bugs. Which, really, compromises our other sprints. So we get kudos for meeting our planned project tasks but then we have to overlook all the quality degradation we introduced.

 

So what usually happens? You might have guessed it: the bugs (assuming they are not "serious enough") are put on a backlog for later fixing. Which seems fine until you realize "later" rarely comes because there are always more features to deliver and more bugs introduced.

 

This spiral is absolutely the most pernicious problem in the development industry (again, game or otherwise). It's the eternal conundrum of the industry: the friction of a product team that needs to get more features out to bring in continued revenue (external qualities) and a development team that (usually) wants to provide solid, cohesive, refactored code that is easy to maintain, extend and scale (internal qualities). I leave it up to personal opinion as to which side is Empire and which is Republic in this context!

Edited by Kryptonomic
Link to comment
Share on other sites

An interesting tidbit, perhaps, is that sometimes the testing done on development shards will reflect the abilities working just fine. It's when the functionality gets put onto production instances that things start to go a bit haywire. This is why public test servers are so important, assuming the public test server has good enough fidelity with the production environment. (Ideally, they are equivalent to the production server in terms of configuration and load possibilities.) And, of course, assuming that the development team is aware of and listening to feedback from those participating in the public testing.

 

I guess it's fair to make the assumption that they do not listen. I've not participated in any public testing since 2.0, but back then I did and I tried to provide the best feedback I could by providing screenshots and even recorded material. My focus at that point was the changes that were being made to the Balance/Madness specialization for Shadows/Assassins and I wasn't alone in criticizing the new changes. But 2.0 launched with nary a change (I can't say for sure if there were no changes at all, but the specialization was underperforming for many months).

 

However, with regards to Phantom Stride and Holotraverse; they should have had the evidence from the live servers that, perhaps, these abilities won't work as they want them to. Exfiltrate has been buggy in PvP since it was first introduced, and later on (in 4.0 I believe) they introduced the abilities Blade Blitz and Mad Dash; abilities that have almost an identical function to Exfiltrate (except they cover more distance and do damage). These abilities are prone to cause characters to disappear as well, just like Exfiltrate. With the way the engine works they ought to reduce movement abilities like these and try to slow down the pace of PvP... but, the damage has already been done. I don't know what the backlash would be if they removed the dysfunctional abilities.

Edited by Majspuffen
Link to comment
Share on other sites

More than likely this was simply a case of the work they committed to having done in their equivalent of the development sprint. If internal project commitments are made, it's considered problematic if those commitments aren't met.

 

For example, if work has been planned out for various sprints, then not reaching each sprint goal puts upcoming sprints in danger of not being able to achieve their goals because now work from the previous sprint has to be incorporated. And so on and so forth.

 

Now, of course, it's still possible to ask: "Well, then why was the Tier 5 Gear upgrade allocated for the same sprint as the Ossus content?" That we can't have an answer to without more insight into the project discussions at the time and the items that are planned for their upcoming sprint cycles.

 

One other note that may be of interest for those who don't work in the development industry (game or otherwise): lots of bugs are a huge friction on team velocity. (Team velocity being the time it takes for the development team to deliver features of an agreed upon quality within an agreed upon timeframe.)

 

So that's the interesting part of the development industry: sometimes we rush to meet our commitments for sprints -- but introduce tons of bugs. Which, really, compromises our other sprints. So we get kudos for meeting our planned project tasks but then we have to overlook all the quality degradation we introduced.

 

So what usually happens? You might have guessed it: the bugs (assuming they are not "serious enough") are put on a backlog for later fixing. Which seems fine until you realize "later" rarely comes because there are always more features to deliver and more bugs introduced.

 

This spiral is absolutely the most pernicious problem in the development industry (again, game or otherwise). It's the eternal conundrum of the industry: the friction of a product team that needs to get more features out to bring in continued revenue (external qualities) and a development team that (usually) wants to provide solid, cohesive, refactored code that is easy to maintain, extend and scale (internal qualities). I leave it up to personal opinion as to which side is Empire and which is Republic in this context!

 

I don't know where you work or got your knowledge, but I sure appreciate some actual realistic explanation behind how the development industry works. No one has ever really gone into the details on how and why here on the forums and that has lead to so much speculation and debate.

 

Sure, we have tons of guesses, even some more educated than others but I feel like your explanations are completely accurate and depict exactly how this field works. Everything you explain confirms some of the guesses and explanations many of the forum posters have had, and then some of the information you have shared fills in a lot of grey areas where only someone who works in this exact field could have such insight.

 

I have to note I find it really interesting that many people who love to oppose other viewpoints here on the forums for the sake of adding "other perspectives" have not touched this thread or responded one time to Kryptonomic's posts.

 

Basically all the theorizing arguments as to "why and how" have found no home in a thread where someone speaks from what appears to be a very direct place in the development field, the very place we often are debating about when concerning the "why and how" of BW/EA's choices regarding the direction of the development of this game.

 

I imagine when it's truth, there simply are no counter points hence no debating or "difference in opinion" to be had.

 

Thanks for your perspective Krypto! That's why these forums are still worthwhile reading from even if you stop playing the game. You never know what you might learn.

Link to comment
Share on other sites

I don't know where you work or got your knowledge, ...

 

I appreciate the sentiment and to respond to this one point, it probably would help for me to set some context. I don't mind people knowing who I am. Here's my LinkedIn profile. During my time at TesterStories (which is my blog and my company) I have worked with various companies, including many game companies on a contract test basis, usually on what's called Story and Narrative Experience Testing.

 

I'll re-mention here a few blog posts I did on SWTOR: those are here and here and here and, finally, here.

 

I mention some of this because I originally went on these forums by my actual name. My tag name was "JeffNyman" here. The Bioware Austin community team made me change that when they felt I was misrepresenting some of my background with this game and with Electronic Arts in particular. This was even after the fact that I was able to provide a signed non-disclosure which not only backed up my claims but also showed I wasn't violating anything.

 

In fact, my non-disclosure contract stated that I had to be on the forums with my actual name, partly so any infractions or violations could be easier to spot. Thus I was actually forced to violate one part of my non-disclosure by the community team! The ironies of the industry, to be sure.

 

I bring all this up not because I'm trying to make the community team look bad at all. Things happen. I totally get it. I'm mentioning all this in the spirit of what I hope is transparency.

 

So ... if you'll forgive a bit of on-topic-but-sort-of-off-topic, one thing I've said a few times but didn't really elaborate on was an orchestrator. Maybe some folks are interested in what that is in the context of MMO testing? If not, now would be a great time to stop reading!

 

Orchestrating MMO Tests

 

Orchestrators are very common in MMOs but, of course they go by many names. One company, in fact, created an automated tool, actually calling it Orchestrator. Think of them as a test harness of sorts. The basic idea is that you have fixtures and factories. The factory instantiates a particular fixture. So an example of a fixture might be, say, "a character with given stats at a certain point in story progression with certain database flags set." Factories can instantiate that fixture, sometimes adding variations.

 

The orchestrator allows multiple of these fixtures to be called up (by factories) in order to apply a massive series of test conditions, which are in turn parameterized by data conditions.

 

As just one example, this is how you can do data-driven testing around balancing classes. Balancing classes, from a programmatic perspective, all basically just boils down to relatively simple math behind the scenes. What makes it complicated are the number of conditions in which that math can operate. The more classes you have and the more abilities, that difficulty scales accordingly.

 

Thus all testing for MMOs is just that: an economy of scale. (That's actually the case for all games, but MMOs display it the most.)

 

As one example that is very common: if your architecture is tightly coupled, you will likely never be able to separate balancing for PvE as distinct from balancing for PvP. But an orchestrator can help you with that because in the context of orchestration you have what amount to PvP data conditions and PvE data conditions. Those are then input to a series of test conditions, where the test condition would be say, a Sith Inquisitor, Lightning Discipline teeing off against a Sith Warrior, Juggernaut Discipline.

 

If you don't have an orchestrator, it means your testing can't be very quick as a general rule. That means even if you get feedback from your testing, you may not be able to implement the changes suggested by it because if you do make those changes you have to run the test suite again -- with all permutations -- to validate. But, remember, the test suite already takes too long! So running it multiple times is often problematic. This assumes, of course, that you fully trust the results of your test suite in the first place.

 

This is where orchestrators allow something called mutation testing, however, where you only mutate certain aspects (say, the damage on some abilities) but the rest are cached and so when you run the test again, all that gets sought are variances based on the mutants. Orchestrators would run these tests in parallel and, keep in mind, don't require the UI. That's a large reason why they are so fast. (This is why damage from a number perspective and animations at the UI level can often seem a little disconnected.)

 

Beyond combat dynamics, this also lets you look for things like story and progression flags that are being modified when they should not be or that aren't being modified when they should. These are often the most interesting because the web of connections between story elements can be a little tangled. Consider the male Smuggler. Did you romance Akaavi? Risha? Or both? And if it was both, did you get the "love triangle" dialogue? And, if so, what did you do?

 

We've recently seen a variant of this bug where the flags are getting misplaced regarding whether you chose Zash or Khem Val, based on events in the Nathema Conspiracy. But, of course, there are many others that people can point to.

 

A most pernicious problem is those flags and mappings I've mentioned a couple of times. One of the most egregious of those shows up with the Taris Bonus Series for the Republic. But this post has gotten pretty long so I'll stop now!

Link to comment
Share on other sites

×
×
  • Create New...