61 replies to this topic

[Endgame124]

Are you doing YOUR part? I honestly and truly recommend a fresh windows install (use acronis to take an image after the install so you can roll back easily). I had a hell of a problem installing MWO back when the MWO beta started. After about 3 days of screwing with stuff, I found the problem to consistently be Star Wars, the Old Republic. MW:O would not install if I had previously run SW:ToR. After I got MW:O running, it would play fine till I launched SW:ToR - if I had played ToR before playing MWO, I would get bugs galore. I reinstalled windows, and ONLY have Team speak, Security essentials, Chrome, and MWO installed, plus all my updated drivers. I rarely have any bugs. About 2 months ago, I started getting a rash of bugs - I rolled back to my Acronis Image, reinstalled MW:O from scratch and have been mostly problem free since.

[Divine Madcat]

Endgame124, on 17 April 2013 - 08:48 AM, said:

Are you doing YOUR part? I honestly and truly recommend a fresh windows install (use acronis to take an image after the install so you can roll back easily). I had a hell of a problem installing MWO back when the MWO beta started. After about 3 days of screwing with stuff, I found the problem to consistently be Star Wars, the Old Republic. MW:O would not install if I had previously run SW:ToR. After I got MW:O running, it would play fine till I launched SW:ToR - if I had played ToR before playing MWO, I would get bugs galore. I reinstalled windows, and ONLY have Team speak, Security essentials, Chrome, and MWO installed, plus all my updated drivers. I rarely have any bugs. About 2 months ago, I started getting a rash of bugs - I rolled back to my Acronis Image, reinstalled MW:O from scratch and have been mostly problem free since.

My install is a clean Win 8 install I made in Feb. As it is, it is silly to blame machines like this, when even PGI has acknowledged the bug..

I agree though,the internal testers continue to drop the ball. The game should have never rolled with the gnarly graphics bug; it makes me wonder if anyone actually tested the game at all....

[Loler skates]

Rizzelbizzeg, on 17 April 2013 - 07:25 AM, said:

<--- non-game software dev

Beta is beta.

LOL at OP not knowing what "patch" means

.....
Troll and liar to boot.

[Divine Madcat]

Loler skates, on 17 April 2013 - 09:00 AM, said:

.....
Troll and liar to boot.

No disagreement, but lets not feed the trolls if possible..

[Sephlock]

They probably went OB in order to bring in more funds so they could hire more people. That's my theory anyway .

[Rizzelbizzeg]

Loler skates, on 17 April 2013 - 09:00 AM, said:

.....
Troll and liar to boot.

.....
QQ bear and debbie downer to boot

Divine Madcat, on 17 April 2013 - 09:18 AM, said:

No disagreement, but lets not feed the trolls if possible..

Nom nom nom

Edited by Rizzelbizzeg, 17 April 2013 - 09:24 AM.

[LiminalSpace]

Mercules, on 17 April 2013 - 08:26 AM, said:

My company still has bugs in their code that has been around for 4 years... 4 YEARS!

The issue though is beyond their control in a piece of code that comes from another company to interact with their hardware. We can't alter the code because of contracts so we have built things in around it to lessen the impact of it, but it is still there. PGI is building off a game engine with limitations and probably undocumented issues in that system too. I expect bugs especially with a bi-monthly development run.

Hell, we have bugs that are over 15 years old. They will basically *never* be fixed unless one of two things happens:
1) An important customer throws a fit about it. We do business software, so our relationship with customers is different from PGI's.
2) A bunch of customers start reporting the issue (and that typically will only happen if some larger business condition changes within some industry, and suddenly everyone is using a part of the system that few used previously).

Some defects are simply never important enough to fix (in preference of other defects or features -- whatever else the development team spends time on). It may suck for a few people who have truly awful issues that are either very narrow cases or could be solved by something like a driver or hardware update, but that's life.

To the OP: regression testing with automatic tests does not catch everything, or even close to everything. To do so, you would have to write a test for *every possible scenario.* This would take one or more orders of magnitude more time than writing the original code itself. For a game, that is a colossal waste of time and effort. This isn't a heart monitor or air traffic control. The same is true for non-automated regressions (i.e., QA analyst excersizing the functionality), but worse.

Plus, there are things that automatic tests are better and worse at. They are a part of an overall testing effort that goes into software, and all software balances level of testing against consequences of defects. There is no QA process that can eliminate all bugs. There is no QA process that can eliminate even all major bugs. QA processes merely reduce the likelihood of bugs, and only insofar as they can test.

And remember: THERE IS NO SUBSTITUTE FOR 50,000 PEOPLE USING THE SYSTEM IN NORMAL COURSE OF PLAY. No "testing" can possibly simulate that. There WILL be bugs, and there WILL be big, problematic bugs. It's all a question of how often they happen, how severe they are, and how PGI handles them. So far, with only a few months under my belt, I think PGI can do a lot better with its testing, and it appears they are aware of that. While some of the issues now are annoying, they are not enough so that I don't want to play. I'm content to see what happens over the next few months.

[Wrenchfarm]

Excellent post OP, too bad it will fall on deaf ears.

"Try a clean install!" "Every product has bugs, stop complaining."

Ugh. I don't even know how to interact with these people anymore. I'm not sure if they just feel the need to defend PGI/the game because they put money into it, or if they really have been whipped by shoddy products and lowered expectations over the years that they are genuinely fine with a product rife with major long-standing problems that have failed to be addressed.

[Merky Merc]

I can play the game again.

I feel like that makes this patch.. a patch.

[Divine Madcat]

Liminal

You are correct, for a game this size, fully automated would be a pain. But good regression testing always includes the user, commonly interacting with the UI. For the game, a good start would be starting the game 10 times, ensuring it starts right. Try all the chat functions. Make sure weapon grouping works. Make sure movement works, etc. Even basic regression testing would have uncovered many of the bugs we have now.

I know PGI is taking on more testing (that should have already been there). So now I implore them to step up to the plate, and have the courage to take a month, and just fix bugs, and get the game back under way..sigh....

[l4Dl]

Divine Madcat, on 17 April 2013 - 05:24 AM, said:

And thus my rant. Things that were once working, are now working worse, or are broken
- The backspace key. Working for months.. now, you can't hold it down. A small issue, but an obvious one.
- Hud Bugs. We went from a pretty much working hud, to one that breaks several times a day. While old bugs were squashed, we got new ones (no map rotation, no IFF, no hud at all).
- Performance issue. One day, people were using a specific nVidia driver just fine, then bam... bugs galore. Mind you, the bug persisted on other drivers, but that makes the point even more.

Couldnt agree more. Patches are a bit "messy" latley.

- Hud/map bugs are still here
- Scoreboard isnt sorted in any order (how did this get past QA? honestly)

And the backspace key, christ, i even gave them the code in C++ and sent it to the devs, was a 5 minute job, 1 minute to add to the game.
http://mwomercs.com/...pace-bug-fixed/

Seems every game dev team works differently, or, the one at PGi is not being controlled correctly and it shows ("do a bit here, do that, break this, maybe fix that")

[Divine Madcat]

Well, I certainly would not be surprised if alot of the horrible directions come from IGP, as they no doubt have a strong influence. I just wish somebody would getthe message through that bugs piling up only hurt their bottom line..

[IceSerpent]

Wrenchfarm, on 17 April 2013 - 09:30 AM, said:

Excellent post OP, too bad it will fall on deaf ears.

"Try a clean install!" "Every product has bugs, stop complaining."

Ugh. I don't even know how to interact with these people anymore. I'm not sure if they just feel the need to defend PGI/the game because they put money into it, or if they really have been whipped by shoddy products and lowered expectations over the years that they are genuinely fine with a product rife with major long-standing problems that have failed to be addressed.

/signed.

The sheer idiocy of some people is mind-boggling - anybody with at least two brain cells stil firing would go straight from "clean install helps" to "there gotta be a bug in the patcher", while certain individuals somehow manage to equate "clean install helps" with "the product is bug-free and we are good to go". I won't even comment on "it's OK to have bugs because every product has them" idea.

[Endgame124]

l4Dl, on 17 April 2013 - 09:52 AM, said:

[/size]
Couldnt agree more. Patches are a bit "messy" latley.

- Hud/map bugs are still here
- Scoreboard isnt sorted in any order (how did this get past QA? honestly)

And the backspace key, christ, i even gave them the code in C++ and sent it to the devs, was a 5 minute job, 1 minute to add to the game.
http://mwomercs.com/...pace-bug-fixed/

Seems every game dev team works differently, or, the one at PGi is not being controlled correctly and it shows ("do a bit here, do that, break this, maybe fix that")

Scoreboard is sorted by lance... Some bugs aren't actually bugs if you actually look at the results every now and then, huh?

[Homeless Bill]

My guess is that they want to, but the publisher demands new content take priority over stability until near launch. I wouldn't hold your breath. Enjoy it for what it is now (when you can) and hope IGP gives them a month to kill bugs before launch.

Because if this game launches in a state similar to what we have now, they're going to get panned in all the reviews, the new playerbase they hope to attract will laugh and eject, and the Founders will be left feeling even more ****** than many of them do now.

[Divine Madcat]

This is the other thing that bugs me..

At least in much of the public eye, we are already launched. Yes, I fully understand the devs call it beta still... but to Joe schmoe, it is as released as it can be (You can buy things with money, you can play all you want... may as well be released).

As such, like it or not, lots of potential players are already forming opinions, and i hate to think what news guys starting this month think..

[LiminalSpace]

Divine Madcat, on 17 April 2013 - 09:46 AM, said:

Liminal

You are correct, for a game this size, fully automated would be a pain. But good regression testing always includes the user, commonly interacting with the UI. For the game, a good start would be starting the game 10 times, ensuring it starts right. Try all the chat functions. Make sure weapon grouping works. Make sure movement works, etc. Even basic regression testing would have uncovered many of the bugs we have now.

Which chat functions? Sure, the obvious ones -- Y = team, T = all, U = lance. Some more you could think of, but the chat functions appear to be fairly broad, including things like "Ctrl-A" for select all, etc. Which ones should be tested? How many test cases do you want to write? Believe me, of the bugs in the game right now, the lack of backspace in chat is my number 1 pain (not the worst bug, just the one that annoys me most often), but writing a test to make sure that the backspace key works in chat seems like a huge waste of time (though now that there's been a bug with it, they certainly should when they fix it to protect against regressions). Should they do Caps lock, num lock, shift, make sure each key on every keyboard can be typed, etc. Even simple things like chat can require fairly massive numbers of test cases if you want full coverage.

Quote

I know PGI is taking on more testing (that should have already been there). So now I implore them to step up to the plate, and have the courage to take a month, and just fix bugs, and get the game back under way..sigh....

Or, perhaps they would like to see how their new automated testing efforts pay off before they do something that could push the game's release back by a month. And bear in mind that there are many players out there who do *not* want PGI to take their foot off the gas for a month. They want missiles HSR, CW, lobby, UI 2.0, and many of the other things PGI intends to deliver. I'm not saying PGI shouldn't spend more time bug bashing, or that taking a month off to do that exclusively isn't necessarily the best solution.

[Divine Madcat]

LiminalSpace, on 17 April 2013 - 01:32 PM, said:

Which chat functions? Sure, the obvious ones -- Y = team, T = all, U = lance. Some more you could think of, but the chat functions appear to be fairly broad, including things like &quot;Ctrl-A&quot; for select all, etc. Which ones should be tested? How many test cases do you want to write? Believe me, of the bugs in the game right now, the lack of backspace in chat is my number 1 pain (not the worst bug, just the one that annoys me most often), but writing a test to make sure that the backspace key works in chat seems like a huge waste of time (though now that there's been a bug with it, they certainly should when they fix it to protect against regressions). Should they do Caps lock, num lock, shift, make sure each key on every keyboard can be typed, etc. Even simple things like chat can require fairly massive numbers of test cases if you want full coverage.


Or, perhaps they would like to see how their new automated testing efforts pay off before they do something that could push the game's release back by a month. And bear in mind that there are many players out there who do *not* want PGI to take their foot off the gas for a month. They want missiles HSR, CW, lobby, UI 2.0, and many of the other things PGI intends to deliver. I'm not saying PGI shouldn't spend more time bug bashing, or that taking a month off to do that exclusively isn't necessarily the best solution.

The new features are great, but is the game is as broken as the last few features.. who cares?

Yes, I concede that there could be a huge number of cases.. and no, they can't all be tested. But given what we have been seeing the last few patches, it is obvious we are missing even basic testing and testers.

As it is, reasons aside, we have to deal with what we have now; that is a slew of bugs. It would be in everyones best interest to stop and fix things before it gets more out of hand...

[LiminalSpace]

IceSerpent, on 17 April 2013 - 10:53 AM, said:

The sheer idiocy of some people is mind-boggling - anybody with at least two brain cells stil firing would go straight from "clean install helps" to "there gotta be a bug in the patcher", while certain individuals somehow manage to equate "clean install helps" with "the product is bug-free and we are good to go".

Who is saying that?

Quote

I won't even comment on "it's OK to have bugs because every product has them" idea.

Since you bring up sheer idiocy, can you explain how it makes any sense to say that it's *not* okay to have bugs when every product has them? It's okay. Really it is. EVERY software product of any appreciable complexity has them. If that's not okay, I shudder to think about your opinions on grammar.

Or perhaps you are trying to say that some subset of the current bugs in MWO are "not okay"? If so, then I certainly agree, but you're not exactly winning any insight awards here. PGI themselves have effectively said that they agree with that when they talked about beefing up QA.

To me, when it comes to MWO bugs, the brass tacks here are as follows:
- Will the ongoing improvements in QA (and, hopefully, accompanying code practices) whip things into shape over the next 5 months or so and get them to a state where releases are solid?
- Will they continue to improve their handling of major bugs when they happen? The last couple major issues resulted in fairly quick communication, then hotfixes.
- How quickly will they be able to absorb and train up the people they are hiring (if what I've read here is to be believed about team expansion)? I can tell you from experience that introducing large numbers of new programmers at one time to a project is a very very very difficult thing to manage, and quality regressions are almost assured.

The story on all of those will be told over the next few months.

[jakucha]

The game runs well for me honestly. Hopefully they fix it for everyone soon, but realize not everyone is having so much trouble with it.