207 réponses à ce sujet

[Freskione2]

Good read..people think its a touch of a button now a days and presto, all better and fixed. doesn't work that way.

[Atheosis]

I've done enough work with software to know that it isn't that hard to fix most bugs. The issue is play-testing to avoid introducing new bugs, and certification. All that said, the game being as buggy as it currently is three and a half months after release is nothing but a joke.

[Mgamerz]

Atheosis wrote...

I've done enough work with software to know that it isn't that hard to fix most bugs. The issue is play-testing to avoid introducing new bugs, and certification. All that said, the game being as buggy as it currently is three and a half months after release is nothing but a joke.

You haven't played Bethesda games then, heh.

[Atheosis]

Mgamerz wrote...

Atheosis wrote...

I've done enough work with software to know that it isn't that hard to fix most bugs. The issue is play-testing to avoid introducing new bugs, and certification. All that said, the game being as buggy as it currently is three and a half months after release is nothing but a joke.

You haven't played Bethesda games then, heh.

Of course I have.  And the point I was making would apply to them just as much as Bioware.  Some designers just have serious QA issues.  That isn't a reason to accept such a thing in general.

[rmccowen]

Atheosis wrote...

I've done enough work with software to know that it isn't that hard to fix most bugs.

Fixing, sure! That's part's relatively easy.

Finding bugs, on the other hand...

[elPrimoFilipino]

Atheosis wrote...

Mgamerz wrote...

Atheosis wrote...

I've done enough work with software to know that it isn't that hard to fix most bugs. The issue is play-testing to avoid introducing new bugs, and certification. All that said, the game being as buggy as it currently is three and a half months after release is nothing but a joke.

You haven't played Bethesda games then, heh.

Of course I have.  And the point I was making would apply to them just as much as Bioware.  Some designers just have serious QA issues.  That isn't a reason to accept such a thing in general.

Don't want to derail the thread, but why are Bethesda's games so buggy?  It's their hallmark now.  Short answer would be great.

Great thread by the way.

[Mgamerz]

elPrimoFilipino wrote...

Atheosis wrote...

Mgamerz wrote...

Atheosis wrote...

I've done enough work with software to know that it isn't that hard to fix most bugs. The issue is play-testing to avoid introducing new bugs, and certification. All that said, the game being as buggy as it currently is three and a half months after release is nothing but a joke.

You haven't played Bethesda games then, heh.

Of course I have.  And the point I was making would apply to them just as much as Bioware.  Some designers just have serious QA issues.  That isn't a reason to accept such a thing in general.

Don't want to derail the thread, but why are Bethesda's games so buggy?  It's their hallmark now.  Short answer would be great.

Great thread by the way.

Their open RPG/Sandbox is so complex that there are so many permutations that its easy to overlook a storyline driven bug (a sidequest bug, I guess.)
RPG's are incredibly complex. I remember when I was starting out with game design, a group I knew was trying to make a pokemon game (for PC)... and they found out it was horrendously complex. ME3 is pretty storyline driven, most of it is linear (not sandbox), but the branching paths make a lot of different areas, it's pretty good how they didn't have many storyline decision bugs.

Modifié par Mgamerz, 26 juin 2012 - 07:28 .

[xtorma]

Mgamerz wrote...

Not every patch is going to fix every issue. Though the first patch did not fix the Vanguard glitch they said they would fix, but I'm sure it'll get fixed at one point, better sooner than later.

how long am i supposed to wait? when is the cut off point where i can say to myself...ok they cant fix this. If you give them infinite time, then there is no reason for them to even try to fix anything. how long should i give them to fix the smglm bug 1 month? 2? 6? or do i just have to accept the fact that they may never be able to fix it.

how much time do they need?

[BjornDaDwarf]

xtorma wrote...

how long am i supposed to wait? when is the cut off point where i can say to myself...ok they cant fix this. If you give them infinite time, then there is no reason for them to even try to fix anything. how long should i give them to fix the smglm bug 1 month? 2? 6? or do i just have to accept the fact that they may never be able to fix it.

how much time do they need?

Also, what's the cutoff for acceptable versus unacceptable bugs? 

The OP's original point is basically that bugs are hard to fix and they take time, particularly when you are dealing with consoles.  True, but there are threshholds which are unacceptable in severity of bug, number of bugs and the time to take to fix them.  The OP doesn't really seem to acknowledge that there are any theshholds at all that the community should find unacceptable.

Also, as an addition to my previous post about better communication helping this issue, you can look at the current stickies.  There are 3 stickies covering cheating and 1 for forum behavior.  So basically 4 stickies regarding player behavior.  But not one regarding known bugs/glitches/problems. 

[Mgamerz]

I didn't state any threshhold since it's going to be different for everyone. I'd say about 3 months is as long as a crippling bug can wait. That said, I'm not sure I'd say the vanguard bug is crippling, just 'severe'.
You would have to ask Derek, he's on QA, I'm sure he's read this thread by now

Modifié par Mgamerz, 26 juin 2012 - 08:14 .

[Catastrophy]

I always wondered about this certification for consoles:

Is it a like a fee they impose for offering a product on their platforms? Like a shopowner taking money for putting things on the shelve?

[BjornDaDwarf]

Mgamerz wrote...

I didn't state any threshhold since it's going to be different for everyone. I'd say about 3 months is as long as a crippling bug can wait. That said, I'm not sure I'd say the vanguard bug is crippling, just 'severe'.
You would have to ask Derek, he's on QA, I'm sure he's read this thread by now

But I am asking you, what do you think the threshold should be?  You started the topic and asked us to discuss it.  Some of these bugs are going on 3 months old, others 2 months (since Rebellion release).  Your OP is criticizing of people who complain and demand bug fixes, and a rational explanation of why fixing cames is difficult.  I agree with the soul of your OP, but going on 3 months with one debatably effective patch is really pushing the envelope for a lot of people.  And it's compounded by having no idea how far out into the future it will be before anything gets fixed.  That lack of communication rears it's head again. 

[RamsenC]

Mgamerz wrote...

I didn't state any threshhold since it's going to be different for everyone. I'd say about 3 months is as long as a crippling bug can wait. That said, I'm not sure I'd say the vanguard bug is crippling, just 'severe'.
You would have to ask Derek, he's on QA, I'm sure he's read this thread by now

It's a crippling bug to me

Modifié par RamsenC, 26 juin 2012 - 08:30 .

[dvjm]

I'm sorry, but the original poster strikes me as giving far too many excuses for a job done badly. And yes, I do know what I'm talking about. Some bugs are hard to find, but many aren't. Some are hard to fix, but many aren't. The fact that we still have map glitches after so long, for instance, is completely unacceptable: these are easy to find (especially if you read the forums) and easy to fix. Now, OK, ME3 isn't as bad as some - the patchwork of bugs loosely held together with more or less working code that is Fallout New Vegas springs to mind as a particularly bad example - but honestly large pieces of software can be better than ME3 is.

I also completely agree with Bjorn about communication. An honest sticky thread on the main issues would be easy to do, and a big step forward.

[Mgamerz]

The vanguard glitch has more to do with server-client communications rather than the attack itself. However, they could [might be able to] fix it there was a check for death between it, and either set your position to the start or to the end of where ever you were going. Not sure how difficult that would be since I don't have the source code to ME3 obviously.

Modifié par Mgamerz, 26 juin 2012 - 08:36 .

[justin_sayne1]

BjornDaDwarf wrote...

xtorma wrote...

how long am i supposed to wait? when is the cut off point where i can say to myself...ok they cant fix this. If you give them infinite time, then there is no reason for them to even try to fix anything. how long should i give them to fix the smglm bug 1 month? 2? 6? or do i just have to accept the fact that they may never be able to fix it.

how much time do they need?

Also, what's the cutoff for acceptable versus unacceptable bugs? 

The OP's original point is basically that bugs are hard to fix and they take time, particularly when you are dealing with consoles.  True, but there are threshholds which are unacceptable in severity of bug, number of bugs and the time to take to fix them.  The OP doesn't really seem to acknowledge that there are any theshholds at all that the community should find unacceptable.

Also, as an addition to my previous post about better communication helping this issue, you can look at the current stickies.  There are 3 stickies covering cheating and 1 for forum behavior.  So basically 4 stickies regarding player behavior.  But not one regarding known bugs/glitches/problems. 

On communication, if you go to the forums that address technical issues, I counted 7 stickies regarding bugs and patches, and more have been there in the past.  There have also been stickies about multiplayer specific issues in this forum at various times.  Not to mention BW jumping into user threads about issues to collect more data.  No, they don't announce their patching cadence, but no software company does (except for Microsoft, and even they don't say WHAT will be patched, just that "something" will be patched every week).  For those that have paid attention, though, you will see that the game has been out for 3 months and we are at patch level 1.03.  Hmmm...I wonder if those numbers are related?

Since you brought up the idea of "acceptable severity", I thought I would share a sample severity ranking:

SEV 1 - System will not function. (I.e. crashes on start)
SEV 2 - System only function in degraded mode or for short duration. (I.e. slows down to unusable or crashes after a time)
SEV 3 - Critical workflow paths cannot be followed. (I.e. engine works fine, but can't complete main tasks)
SEV 4 - Logic error (I.e. host migration error)
SEV 5 - Cosmetic (I.e. Falls off map, wrong graphic, weapon shoots wrong projectile, etc.)

I applaud BioWare QA that we did not see any SEV 1, 2 or 3 issues (OK, well, except for the SEV 3 "online pass" issue after a patch), and only a few SEV 4.  Plus, I am certain that the number of SEV 5 that got out the door pales in comparison to the ones that didn't.

[Mgamerz]

justin_sayne1 wrote...

BjornDaDwarf wrote...

xtorma wrote...

how long am i supposed to wait? when is the cut off point where i can say to myself...ok they cant fix this. If you give them infinite time, then there is no reason for them to even try to fix anything. how long should i give them to fix the smglm bug 1 month? 2? 6? or do i just have to accept the fact that they may never be able to fix it.

how much time do they need?

Also, what's the cutoff for acceptable versus unacceptable bugs? 

The OP's original point is basically that bugs are hard to fix and they take time, particularly when you are dealing with consoles.  True, but there are threshholds which are unacceptable in severity of bug, number of bugs and the time to take to fix them.  The OP doesn't really seem to acknowledge that there are any theshholds at all that the community should find unacceptable.

Also, as an addition to my previous post about better communication helping this issue, you can look at the current stickies.  There are 3 stickies covering cheating and 1 for forum behavior.  So basically 4 stickies regarding player behavior.  But not one regarding known bugs/glitches/problems. 

On communication, if you go to the forums that address technical issues, I counted 7 stickies regarding bugs and patches, and more have been there in the past.  There have also been stickies about multiplayer specific issues in this forum at various times.  Not to mention BW jumping into user threads about issues to collect more data.  No, they don't announce their patching cadence, but no software company does (except for Microsoft, and even they don't say WHAT will be patched, just that "something" will be patched every week).  For those that have paid attention, though, you will see that the game has been out for 3 months and we are at patch level 1.03.  Hmmm...I wonder if those numbers are related?

Since you brought up the idea of "acceptable severity", I thought I would share a sample severity ranking:

SEV 1 - System will not function. (I.e. crashes on start)
SEV 2 - System only function in degraded mode or for short duration. (I.e. slows down to unusable or crashes after a time)
SEV 3 - Critical workflow paths cannot be followed. (I.e. engine works fine, but can't complete main tasks)
SEV 4 - Logic error (I.e. host migration error)
SEV 5 - Cosmetic (I.e. Falls off map, wrong graphic, weapon shoots wrong projectile, etc.)

I applaud BioWare QA that we did not see any SEV 1, 2 or 3 issues (OK, well, except for the SEV 3 "online pass" issue after a patch), and only a few SEV 4.  Plus, I am certain that the number of SEV 5 that got out the door pales in comparison to the ones that didn't.

SEV2 - The final battle on Crysis. That had so many memory leaks I had 8GB of ram being used by that final battle, and my computer could STILL barely handle it.

[BjornDaDwarf]

justin_sayne1 wrote...

On communication, if you go to the forums that address technical issues, I counted 7 stickies regarding bugs and patches, and more have been there in the past.  There have also been stickies about multiplayer specific issues in this forum at various times.  Not to mention BW jumping into user threads about issues to collect more data. 

None of the stickies currently in the tech support areas address known current bugs, which is specifically what I thought would help cut down on the number of ME3 MP threads complaining about bugs.  And, as I said in my big post earlier on this, communication has been improving, but is very inconsistent.  There are some bugs that have never been acknowledged.  

The OP, and may of the repliers, come from programming/CS background.  For full disclosure, I come from a communications background (used to be journalism, now PR/customer service).  The way that so many companies struggle to communicate effectively in the modern era is a constant irritant for me. 

In the tech support forums, Priestly actually does have a good stickie about the process of bug fixing.  I'm a little surprised it isn't also stickied in the MP thread.  I think we could give up one of the three cheating threads to add the bug/patch one.  But I do want to highlight his very first point from that:

Chris Priestly wrote...

• Everything starts with help from customers like you who report issues here on the BSN or to our EA Custmer Service people.

The OP is being critical of people harping about issues.  And lord knows that the general tone of the BSN could be a lot friendlier some days.  But all those threads are serving a purpose that BW asked for.

Modifié par BjornDaDwarf, 26 juin 2012 - 08:56 .

[justin_sayne1]

Mgamerz wrote...

SEV2 - The final battle on Crysis. That had so many memory leaks I had 8GB of ram being used by that final battle, and my computer could STILL barely handle it.

Ouch!  I had one like that on Dragon Age: Origins.  After visiting the Dwarves, load times would get progressively longer and take more and more memmory.  I was contemplating putting in 64GB ram just to get decent load times.

[justin_sayne1]

BjornDaDwarf wrote...

The OP is being critical of people harping about issues.  And lord knows that the general tone of the BSN could be a lot friendlier some days.  But all those threads are serving a purpose that BW asked for.

Maybe it is the lack of "tone" inherent in a textual form of communication, but I had not read his OP as being critical of anyone.  Also, there is a difference of contributing to a thread about a known issue, or starting a new thread if you find one, versus "I am starting a new thread because the issue I started a thread about 5minutes ago is not fixed yet, so BW sucks" which seems very common on the MP forum these days.  I saw the OP as addressing the last with a bit of "insider info".

[Mgamerz]

Not trying to be critical. But most people don't know what it's like to do coding work. I thought I'd provide a little bit of insight since BSN is become pretty toxic.

[xtorma]

Mgamerz wrote...

I didn't state any threshhold since it's going to be different for everyone. I'd say about 3 months is as long as a crippling bug can wait. That said, I'm not sure I'd say the vanguard bug is crippling, just 'severe'.
You would have to ask Derek, he's on QA, I'm sure he's read this thread by now

there has to be an average time or no one could get a business plan past the bank, or a board. Why is it different for everyone? I would assume it's because different teams have different skill levels. If this is the case , at what point to we give up and just accept that it wont be fixed?

Obviously no one outside of the company can answer that, in fact , according to you , no one within the company can answer it. 

BTW  lol at the vanguard bug not being crippling. Are you basing that on the fact that we have so many classes to choose from , or are you basing it on the fact that it's ok to fall through the map once every 3 games?

[Ninja Stan]

There are many factors that also prevent bugs from being fixed. Fixing bugs and releasing patches takes more than zero time and effort, so that time and effort have to be spent wisely. Do you fix a couple of giant bugs, or knock out as many tiny bugs as you can? Fixing the giant bugs is super noticeable, but fixing a ton of tiny bugs is also noticeable, just less so.

Bug priority is a big thing. Yeah, you can fix a couple of giant bugs, but how many people does it affect? How easy is it to reproduce? If it's a hugs bug, but only affects a few people, the priority of that bug might be reduced in favour of a smaller bug that affects everyone. Of course, giant bugs that affect everyone will usually have high priorities.

One factor that many people will likely be unaware of is stability. Sure, there may be a big bug that's affecting a bunch of people, but if it's in a system that has a lot of dependencies or is unstable to begin with, messing with the code in that area is like taking a Jenga block out of a section that's already got some blocks removed. Sure, you might fix it without affecting any other system, but then, the entire thing could come crashing down. Of course, once again, huge bugs affecting everyone (such as game crashes, save corruption, and plot states) are usually given priority regardless of how unstable the system is (which may be a bug in and of itself) or how long it takes to fix.

Many people get rather impatient waiting for fixes, but remember all of the steps that need to be followed in order for a fix to appear for you. First of all, the bug needs to be reproduceable. The easier it is for a dev team to reproduce a bug, the easier it'll be to find the root cause and determine whether a fix is feasible. As implied in my previous paragraphs, not all fixes are feasible, no matter how you as a gamer feel about it. Once a bug is reproduced and fixed, it has to be tested a lot. This is to ensure the bug doesn't cause other problems. The more obscure the bug, the harder it is to test. And QA needs to test the fix as cleanly as possible, so if a bug only occurs at the end of the game, QA might have to play through the whole game without cheating any number of times in order to ensure it's fixed. (Remember this the next time you have any kind of "completionist" achievement in a game. Someone had to test that in-game without cheating. Multiple times. Likely for months.) The more dependencies a system has, the more QA has to test, in order to ensure those systems are also not negatively affected by the fix.

Finally, once the fix has been verified in a build--and note that many different fixes are likely implemented at a time in new builds--that new build itself has to be tested. Essentially, the game itself (with the fixes) has to be re-tested.

Done, right? The build's good, which means you get to play the fixed game. Unfortunately, there are still a couple of things left to do. Important things.

The fix had better not be affected by VO or text, because there is more than one version of the game out there, and not all of them display things the same way. This is why patches don't usually include text fixes, translation changes, or wholesale changes to GUI size or layout. You don't want to render the German text unreadable, say, or make the player download hundreds of MB worth of VO or animation files for a patch.

Secondly, the patch installer has to be built and tested. For every platform. For every language. For every version. Not everyone will be installing the patch from PC region 1 ver. 1,01. What about 1.02, 1.03, 1.04, etc.? What about Steam or other digital versions? The Eastern Europe SKU? And consoles? All of those have to be able to patch to the correct version.

And finally, the whole thing has to meet each console manufacturer's certification standards. All product releases on Xbox and PS3 have to work a certain way, and they have to meet certain standards of presentation and reliability. If a developer or publisher fails to meet a certain number of those standards, or miss even one of the must-have standards, the product cannot be approved for release. BioWare, for example, has an entire (tiny) department devoted entirely to knowing about, testing, and documenting the state of these standards.

So if you experience an issue in a software product, before making a statement like "why didn't QA catch this?" please note that QA is just one step among many between unintended behaviour being noticed and a user running a fixed version of the product.

[Kasrkin]

Toxicity; granted.

3 months of live "playtesting", check.

Balance issues aside, there are several bugs/glitches that need to be *recognized* by Bioware as recreatable/repeatable by them, and acknowledged as existing.

A "Current Known Issues" thread/sticky. That in and of itself would defuse a lot of the frustration that people have in regards to MP.

Not a review of 3 million lines of code. A thread. The absence of said thread speaks volumes to me.