Basing your entire argument on a single post made in a single thread on a single forum seems a little shaky.

What is more tricky to do is to replicate the precise conditions that causes specific issues to occur

they can't test is because CCP don't have SponkSponkSponk's PS3, or Mango Dee's PS3, so they can't do a fair test, it is not CCP's problem

for the lag
-it may cause by your PS3, don't put it behind your TV if you do, heat of TV can make PS3 lag, and place your PS3 at some place cooler or wind flows, or find a mini fan to cool your PS3, at least this work on my PS3
-it may cause by worge battle server region, worge battle server region makes me lag so much that I flying around the map, so check your battle server region is corret or not

you sir are an idiot.

If it makes you feel any better, I run through the technical support/bugs section daily to find and record bugs that people report so that I can upload them to youtube and help narrow them down.

There's an OLD catch in the world of Videogames, born with Ultima Ascension.

"IF the the game is Huge, Huge are the Bugs"

I'm not a CCP fanboy, but we must say that QA can't do all the test. Too many variant, too many thing a player can do, to many conditions.

Otherwise you don't understand why TES or Fallout always come out with huge bugs. And we are talking about a multimillion company making multimillion game.

But some concern about QA is coming from me: why the HMG fix took SOOOOOOOOO long, when all you need to do is try it?

God bless Ghab, savior of the heavys

Oh brother, the power of anonymity at work.

Probably they are few, so their play test time is limited.

Probably they have been busy trying the new scout and heavy, to not make them OP.

Ghost, Have you ever worked on a software development team? Because your questions/attitude suggest you have no idea the complexity associated with debugging or the realities associated with development and testing of code -- not to mention the title of your thread is flat out incorrect....Leadfoot

Bug removal in coding can be an extremely frustating activity, an entire feature could produce a stack overflow if you didn't quite get the terminating conditions on a recursive algorithm just right. Or calculations incorrect because you made a typo or accidentally transposed a character or fatfinger an extra "." or number.

Sometimes bug-hunting and removal is the greatest detective activity you can play, and the best help you can get is quantity. That is something that can be unfeasible in all but the supergiants of game development companies, thousands of people just playtesting. It gets very expensive. However, Mmre people giving their insight, more people submitting videos with a different perspective, it all helps to find that mission condition in the recursion. or that extra ".".

Find a few more examples of this occuring, otherwise this is just an edge case where the Q/A team are having difficulties because of the multiple variables occuring and haven't got a high staffing level or the time to account for all these variables and the players have difficulty because they don't have access to replicate their bugs in a controlled environment.

This thread is so ******* painfull to read. Ghost how the **** do u think make the game and complie the code. The very nature of programing means u can test the dam thing.

Programing games is one of the hardest things to do. Thats why games programming corses allow one of wtf bugs. Because its complety possable u simple never encoutered it before.

I new one guy whos code failed due to chipset of the 1 computer of a lap of hundered being diffrent. His face was awsome when his (eventual perfect pass) blue screened for the first.

This why programer forgive strange bugs and go irate over blantent ones :-P

Let me know when you make an intelligent post.



That just proves the bugs are difficult to replicate, I can say I have never ran into such bugs and I bet a majority of people haven't either. If you can't replicate it, and no-one has clear evidence it exists then "absence of evidence is evidence of absence"

Devs ask for the videos because if you already have one or know how to replicate it by step-by-step instructions, why duplicate work?

Stumbling across a bug during use is way different than forcing one out. I program industrial programmable logic controllers (PLC) and there are times when bugs only do something if something else is half way through cycle c and only while in subroutine 3. We also program with icons that represent the action and provide feed bad data. It is hard to do like that, I couldn't imagine trying to read all of that.

So what's your point? That some bugs are still in the game because we can't find the environment in which they are caused?

You just proved my point.

No, again, because if you can't replicate it 100% without fail, you haven't found the sole source of the bug and CCP maybe be slightly closer in their search but not by much. Also because we, the players, number in the thousands, and If we say for the sake of argument and numbers CCP has 100 employed playtesters in their Q&A and we have 5000 people playing the same hours as the playtesters work. The players are 50x more likely to find a bug and remember the situation that caused it. If we then remember the playtesters need to be paid, well go with a low conservative estimate of -ú15000 a year. That's -ú1,500,000 a year just to find bugs.

I'm all for bug finding, but I also like to be efficient with what I do. Instead of hiring more people, implement tools which can record a Dust clients last 10(?) minutes of input and varibles, and put estimates at what timing the bug happened, they programmers can run through that timeframe of variables and see if they see anything unusual. I don't know if that's possible, but I'd rather CCP spend their resources and money on finding better ways to gather data when bugs occur, rather than just to lever them out with the brute force of sheer numbers.

What if the sterile test environment is getting rid of those variables in the first place, and therefore the only place that the bug can be found is in the live environment?

And what if we can't send them logs and they are simply working on something to solve such an issue?

Ghost is right, a lot of bugs are seemingly random to us players; forge gun dry fire, RDV that just sits there not dropping the vehicle, RDV that yoyos the vehicle, rail turret installations that are constantly firing, dying from hitting the ground even though the inertia dampener was on, vehicles that exploding the moment they tough the ground after being dropped by the RDV, vehicles that fall through the ground after being dropped by the RDV, tanks turning the wrong way when the turret is turned too, gun that jumps while taking a short fall in ADS, splash damage not applying at all, splash damage applying to some people but not the person standing right there next too them, splash damage being the only thing applied in direct hits, tank reload bug (like the handheld weapon one we used to get), .

Those are just the things I could think of off the top of my head and each one of them randomly happens and there's no way for me to figure out why because I don't have the proper tools for it. CCP says they have a Q/A team, well they better hire more people for it because we don't have the proper tools for the job and we're not beta testers anymore.

I wanted to stay away from posting, but I just have to step in on CCPs side.

To say they don't perform Q/A is bull, and the nature of certain bugs makes them VERY hard to track down.
Yes, they have dev kits, but there are certain aspects CCP simply can't replicate and accurately simulate.
(The entire Matchmaking hick up during the launch of 1.4 is proof of that).

You also have to realice that ALL reported bugs get prioritized with criticality and priority, which depends on the severity, reproducibility and available workarounds.
The rail bug is probably far down in the backlog.

I have to say that taken into account the size of the company, the fact that this is their first FPS, their first Console game, and is currently undergoing a huge reorganization (since Uprising) and lastly a new EP, I think CCP has done outstanding work with the level of progress we have made since Uprising came out.

Dev tools don't actually help you reproduce a problem. They help you once you've reproduced the problem and let you start to zero in on what is actually causing it. I've been a developer for 8 years now, mostly in healthcare which, as you can imagine, is the most anti-bug sector you can be in. If something doesn't work right people start flipping tables.

People will report a bug, but if they can't give us the steps to reproduce it all we've got is "there's a problem". You can try screwing around in the app to reproduce it, but that's just a blind, dumb luck search. Until the developer can actually see it happen on his machine there isn't a damn thing that can be done to fix it. There is no such thing as a tool that reproduces a bug that the steps are not known for.

If the bug happens only on occasion and doesn't break the game, why should it be given such a high priority over things such as more game content and game balance?

If the bug was crippling then we'd be able to reproduce it, and the logs would show clues that lead to the source of the problem. If the problem can't be reproduced then it is, by definition, not crippling.

That tool is what I was talking about when I said "They help you once you've reproduced the problem and let you start to zero in on what is actually causing it". You can't step into the code where the bug is occurring if you can't make the bug happen. Without the steps to reproduce you're not going anywhere.

It's annoying to you, but the game is still running and 99.9% of the other people playing at that moment are having no problems with their gun not firing. You're falling into the user reporting a bug trap of, "It happened to me therefore this is the most important thing ever".

Seriously, go work in software development for a while. It's absurd. Users will report bugs and if they're allowed to rate the severity every last bug will be critical or at the very least moderate, even if it is something that happened once to one user and didn't stop anything from finishing.

They're really not helpful at all until you can observe the problem happening. Quite often a "bug" isn't actually a bug, but a problem with the user's environment. I could start my debugger and run through the app all day long. If I did run into the bug then yeah, I've got what I need to get started. The problem is you're expecting the tools to also help you get through that reproduction step. It doesn't work and it can't work. If someone ever builds a tool that can do that for us then we've basically created skynet and are ready to let autonomous computers solve all the world's problems.

a small-minded and foolish response.

If they really are working hard then how did we get 1.4 to happen lets not venture into the dark days again. It only lasted for a day but CCP should be ashamed of themselves that was terrible. What about the flaylock, the murder taxi, the revival glitch when uprising first came out. Everyone went ballistic on the forums when all of those happened well maybe not the revive glitch but the other two definitely yes. It took 2 months!!!!!!!!! to fix each of those evil things and does CCP even actually care when Flaylocks and murder taxis were a problem? Everyone had really good ideas when those were the big problems and we had to wait through each patch to get these things nerfed and balanced out so I agree with Ghost here even though I've seen him do naughty things COUGH COUGH invisibility glitch COUGH COUGH.